Anemarrhenasaponin I suppresses ovarian cancer progression via inhibition of SHH signaling pathway

Cells and cell culture

The human normal ovarian epithelial cell line IOSE80 and ovarian cancer cell line SKOV3 and A2870 and the murine fibroblast cell line NIH 3T3 were purchased from the American Type Culture Collection (ATCC, USA). SKOV3 cells were cultured in McCoy’s 5A medium (KeyGEN BioTECH, China). IOSE80, A2870, and NIH 3T3 cells were maintained in DMEM medium. The cell culture media were supplemented with 10% fetal bovine serum (FBS, Wisent, Canada), 2 mM GlutaMax™ (Gibco, USA), 1 mM sodium pyruvate (Gibco, USA), and Penicillin-Streptomycin-Neomycin (PSN) Antibiotic Mixture (50 μg/mL penicillin, 50 μg/mL streptomycin, and 100 μg/mL neomycin; Thermo, USA).

Luciferase reporter assay

Luciferase reporter assay was performed to examine the effects of herbal medicines on the transcriptional activity of SHH signaling pathway. The 3T3 cells were treated with icariin (Aladdin, China, 3.93 μg/mL), bupleurum (Macklin, China, 10 μg/mL), berberine hydrochloride (MCE, Australia, 40 μg/mL), usnic acid (Aladdin, China, 20 μg/mL), emodin (Titan, China, 25 μg/mL), betulin (Macklin, China, 18.27 μg/mL), An-I (MCE, Australia, 160 μg/mL), resveratrol (Aladdin, China, 1.3 μg/mL), oleander glycoside (Sigma, America, 10 μg/mL), podophyllotoxin (Aladdin, China, 2.5 μg/mL), schisandrin B (MCE, Australia, 2.1 μg/mL) and shikimic acid (Titan, China, 1 mg/mL) for 24 h, separately. The culture medium was discarded and washed twice with phosphate-buffered saline (PBS). The cell lysis solution was added and then placed on a shaker for 20–30 min to completely lyse the cells and obtain the supernatant. Add 10 μL of supernatant and 100 μL of pre-mixed Luciferase Assay Reagent II to each well of a 96-well enzyme labeled plate, detect the intensity of the luciferase reaction, and record the loading sequence and readings. 12 herbal powders were dissolved in Dimethyl sulfoxide (DMSO) as mother liquor at a concentration of 1 mM and were diluted with cell culture medium to the desired concentration before use.

Cytotoxicity assay for IC50

SKOV3 cells and A2870 cells were inoculated in 96-well plates at a density of 4 × 10³/well and 2.5 × 10³/well, respectively. After incubation for 24 h, 0.35, 1.55, 6.25, 25.00, and 50.00 μM An-I were added to treat the cells for 24 h. 10 μL Cell Counting Kit-8 (CCK-8) (APExBIO, USA) solution was applied to each well and the samples were incubated for 1–4 h. The absorbance at 450 nm was measured by a microplate reader (Tecan, Austria) and the inhibition rate was calculated.

RNA extraction and quantitative real-time PCR

Total RNA was isolated from treated cells using the TRIzol reagent (Takara, Japan), and cDNAs were generated with the HiScript II Q RT SuperMix for qPCR kit (Vazyme, China). RT-PCR System (Roche, Switzerland) was carried out using the AceQ qPCR SYBR Green Master Mix (Vazyme, China) on a RT-PCR system (Roche, USA) with primers. Primer sequences used for real-time PCR were human,the details were as follows:

18S-F,5′-CAGCCACCCGAGATTGAGCA-3′ 18S-R,5′-TAGTAGCGACGGGCGGTGTG-3′

GLI1-F, 5′-AGCTAGAGTCCAGAGGTTCAA-3′

GLI1-R, 5′-TAGACAGAGGTTGGGAGGTAAG-3′

SUFU-F, 5′- ACATGCTGCTGACAGAGGAC-3′

SUFU-R, 5′- CACTGCTGGGCTGAGTGTAG -3′

The expression fold change of genes was calculated by the 2^−ΔCT method as references for normalization. Each target was measured in triplicate.

Western blotting

Treated cells were washed twice with PBS, then lysed in 200 μL RIPA lysis buffer. Protein concentrations were determined using a BCA protein assay kit (Thermo Fisher, USA). Total proteins in each cell lysate were separated by 10% or 8% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. The membranes were then blocked with 5% nonfat milk in TBST (50 mM Tris-HCl, PH 8.0, 150 mM NaCl, and 0.1% Tween-20) for 1 h and then incubated overnight at 4 °C with appropriate dilutions of primary antibodies, including rabbit anti-GAPDH (1:1000, Santa Cruz Biotechnology, USA), rabbit anti-ACTIN (1:5000, Affinity Bioscience, USA), rabbit anti-GLI1 (1:1000, CST, USA), rabbit anti-SUFU(1:1000, Abcam, USA). After incubated with horseradish peroxidase-conjugated IgG (1:5000, Jackson ImmunoReasearch, USA) at room temperature for 1–2 h, the antibodies were detected with an enhanced chemiluminescence (ECL) kit (Bio-rad, USA). GAPDH and ACTIN were used as controls to verify equal protein loading.

Cell proliferation assay

Cell proliferation was assessed with Cell Counting Kit-8 (CCK-8) (APExBIO, USA). SKOV3 cells and A2870 cells were respectively seeded at a density of 4 × 10³ and 2.5 × 10³ cells per well in a 96-well plate and cultured for 24 h. Then SKOV3 cells were treated with An-I (25 μM) or GANT61 (10 μM) for 24 h, 48 h, 72 h, 96 h and 120 h, respectively. A2870 cells were treated with An-I (15 μM) or GANT61 (20 μM) simultaneously. The incubation continued for another 4 h after adding CCK-8 and the absorbance at 450 nm was measured with a microplate reader (Tecan, Austria). GANT61 powder (Selleck, USA) was dissolved in ethanol with a concentration of 10 mM and diluted with medium for experimental usage.

EdU Cell Proliferation Assay Kit (RiboBio, China) was also used to analyze cell proliferation. SKOV3 and A2870 cells were planted on coverslips at a density of 5 × 10⁴ per well in a 24-well plate. After adherence, SKOV3 cells were treated with An-I(25 μM) and GANT61 (10 μM) for 20 h. A2870 cells were treated with An-I (15 μM) and GANT61 (20 μM) at the same time. Then cells were incubated with EdU for 2 h. Apollo staining was applied after immobilized with 4% Paraformaldehyde (PFA) and permeabilized with 0.5% TritonX-100/PBS solution. The nuclei of cells were stained with DAPI. Images were acquired by fluorescence microscope.

Clone formation assay

In vitro tumorigenicity was evaluated by colony-forming assay. SKOV3 cells were treated with An-I(25 μM) and GANT61(20 μM), and A2870 cells were treated with An-I(15 μM) and GANT61(20 μM), respectively. 24 h later, cells were digested, and 1.0 × 10³ SKOV3 cells and 1.5 × 10³ A2870 cells were seeded into 6 cm dishes. Cells were maintained in a humidified 5% CO₂ incubator at 37 °C for 10 days (SKOV3) or 12 days (A2870) with fresh medium replaced every 3 days. Then The cells were washed twice with PBS, fixed with 4% PFA for 10 min, and stained with crystal violet for 5 min. The number of colonies was counted after rinsing with running water.

Transwell matrigel migration and invasion assay

The transwell chambers (24 wells; Millipore, USA) were used for the in vitro migration and invasion assays. After treating SKOV3 and A2870 cells with An-I and GANT61 at certain concentrations for 24 h, the cells were transferred into transwell chambers with or without matrigel (BD, USA). A total of 5 × 10⁴ (SKOV3) or 6 × 10⁴ (A2870) cells were placed in each insert chamber without FBS, while the lower chamber contained 10% FBS. Before being fixed with 4% formaldehyde and stained with crystal violet, SKOV3 cells were allowed to migrate for 11 h and invade for 12 h, and A2870 cells were allowed to migrate for 48 h and invade for 48 h. The cells remaining on the upper chambers were scraped off by Q-tips. The number of cells that had migrated and invaded through the matrigel membrane was counted under a microscope.

RNA-seq data generation and analysis

RNA-seq was generated and analyzed as previously described [21]. Total RNA was isolated with RNAiso Plus reagent (TaKaRa) from IOSE80, SKOV3, A2870 cells treated with DMSO or An-I for 24 h, then purified to meet the following requirements were used in subsequent experiments: RNA integrity number (RIN)>7.0 and a 28S:18S ratio >1.8. cDNA and libraries for each sample were prepared independently, and sequenced on an Illumina Novaseq 6000 sequencer by CapitalBio Technology (Beijing, China).

Statistical analysis

Statistical analyses were performed using the GraphPad Prism Version 9.0 program. Data were presented as the means ± SD The difference between the two groups was analyzed using Student’s t-test. Multiple-group comparisons were determined by one-way ANOVA with Tukey’s post-hoc test. A p-value less than 0.05 was considered statistically significant. Each experiment was repeated at least three times.

Chinese herbal medicines were screened and evaluated for the inhibitory activity of SHH signaling

Many studies have been shown that abnormal signaling pathway of SHH is implicated in the development of several types of human cancer. In order to find potential therapeutic approaches for these cancers, we screened Chinese medicinal herbals to examine their anti-SHH signaling activity by evaluating the expression of its downstream target genes such as GLI1 as well as PTCH1 in the NIH 3T3 cells. The results showed that the mRNA levels of GLI1 and PTCH1 dramatically upregulated after treatment with SHH agonists SAG and the ligand, whereas they were significantly inhibited with Berberine hydrochloride, An-I, Resveratrol, and Podophyllotoxin treatment (Figure 1A and B). To further explore the effects of various herbal medicines on the transcriptional activity of GLI1, which is the major transcription factor to mediate SHH signaling, we next performed an 8 × GliBS-luc reporter assay in NIH 3T3 cells. The results revealed that SHH signaling agonists and the ligand greatly enhanced GLI-luciferase activity, while three herbal medicines, just like Berberine hydrochloride, An-I and Podophyllotoxin repressed the transcriptional activity of GLI1, among which An-I had the most obvious inhibitory effect (Figure 1C). These results implied that An-I has the ability to inhibit SHH transcription, which may be an effective target therapy to be used in the treatment of many cancers.

Figure 1:

An-I has the most obvious inhibitory effect on SHH signaling pathway in NIH 3T3 cells. (A and B) Berberine hydrochloride, An-I, Resveratrol, and Podophyllotoxin decreased GLI1 and PTCH1 mRNA levels. (C) The luciferase reporter gene assay showed that Berberine hydrochloride, An-I and Podophyllotoxin inhibited the transcriptional activity of SHH signaling pathway. Data are mean ± SD from triplicate experiments. *p<0.05; **p<0.01; ***p<0.001; ns not significant (unpaired Student’s t-test). Each experiment was repeated at least three times.

An-I reduced SHH signaling activity in ovarian cancer cells

According to published papers, SHH pathway is aberrantly activated in ovarian cancer, and its abnormal activation can lead to the development of ovarian cancer. To confirm these previously reported findings, we examined GLI1 protein levels and transcriptional activities in ovarian cancer cell lines. Indeed, our results demonstrated that protein levels of GLI1 and the mRNA levels of GLI1 and PTCH1 were dramatically elevated in SKOV3 and A2870 cells compared to normal ovarian epithelial cells IOSE80 (Figure 2A–C).

Figure 2:

An-I kills ovarian cancer cells in a concentration-dependent manner. (A and C) Western blotting and qPCR indicated that the SHH pathway was in an abnormal activation condition in SKOV3 and A2870 cells compared to IOSE80. (D and E) CCK8 assay showed IC50=54.57 μM for SKOV3 and IC50=22.57 μM for A2870. (F and I) Western blotting and qPCR showed that An-I downregulates the expression level of GLI1 in a concentration-dependent manner. **p<0.01; ***p<0.001; ns not significant (unpaired Student’s t-test). Each experiment was repeated at least three times.

In order to identify whether An-I is a novel therapeutic strategy for ovarian carcinoma, we evaluated the inhibitory activity of An-I on SHH signaling in ovarian cancer cells. Firstly, Cell proliferation assay was performed to determine the half-inhibitory concentration (IC50). The results revealed an IC50 value of 54.57 μM for SKOV3 cells and 22.57 μM for A2870 cells, and we used these concentrations for further experiments (Figure 2D and E). Subsequently, we investigated whether An-I regulated SHH signaling activity in ovarian cancer cells. Treatment with the different concentrations of the An-I decreased GLI1 protein and mRNA levels in both SKOV3 and A2870 cells in a dose-dependent manner (Figure 2F–I). Notably, as a potential SHH inhibitor, An-I was more effective than GLI1 inhibitor GANT61 in the same concentration range (Figure 2F–I). Surprisingly, when SKOV3 cells were treated with 75 μM An-I, GLI1 mRNA expression was increased instead, as did GANT61, the exact mechanism of which remains to be explored (Figure 2H).

An-I effectively inhibits the proliferation of ovarian cancer cells

Aberrant activation of SHH signaling promotes ovarian cancer cell growth. We previously showed that the Chinese herbal medicine An-I had an inhibitory effect on SHH signaling activity. Therefore, we examined the inhibitory effect of An-I on ovarian cancer cell proliferation. SKOV3 and A2870 cells under An-I treatment grew slower, as demonstrated by the CCK8 assay. The two cell lines were more sensitive to An-I than GANT61 (Figures 3A and 4A). Since 5-Ethynyl-2′-deoxyuridine (EdU) incorporation is a fast and sensitive method to assess cell proliferation, we also analyzed the relative numbers of EdU-labeled cells with or without An-I treatment. Consistent with CCK8 assay, the proliferation of SKOV3 and A2870 cells was significantly decreased in the An-I treated group compared to the control group (Figures 3B and C and 4B and C). We finally performed a colony formation assay to confirm the above results further. The results showed that cells treated with An-I formed fewer colonies than non-treated cells (Figures 3D and E and 4D and E). These data demonstrated the antiproliferative effects of An-I on ovarian cancer cells.

Figure 3:

An-I inhibits the growth of SKOV3 cells. (A) CCK8 assay showed that both An-I and GANT61 could significantly suppress the proliferation of SKOV3 cells. (B and C) EdU assay demonstrated that An-I could suppress the proliferation of SKOV3 cells effectively, while the inhibitory effect of GANT61 was not obvious. (D and E) Colony formation assay exhibited that both An-I and GANT61 suppressed the clonogenic ability of SKOV3 cells remarkably. *p<0.05; **p<0.01; ***p<0.001; ns not significant (unpaired Student’s t-test). Each experiment was repeated at least three times.

Figure 4:

An-I inhibits the proliferation of A2870 cells. (A) CCK8 assay showed that both An-I and GANT61 greatly inhibited the proliferation of A2870 cells. (B and C) The EdU assay showed that both An-I and GANT61 could seriously depress the proliferation of A2870 cells. (D and E) Colony formation assay showed that both An-I and GANT61 were capable of inhibiting the clonogenic ability of A2870 cells effectively. *p<0.05; **p<0.01; ***p<0.001 (unpaired Student’s t-test). Each experiment was repeated at least three times.

An-I suppresses ovarian cancer cell migration and invasion

Aberrant activation of SHH pathway also has an effect on the invasiveness enhancement of ovarian cancer cells. The migratory and invasive capacities of SKOV3 and A2870 cells were examined with transwell assays with or without An-I. As expected, An-I significantly inhibited the migratory and invasive abilities of SKOV3 and A2870 cells (Figure 5A–F). Furthermore, epithelial-mesenchymal transition (EMT) is a trans-differentiation process for epithelial cancer cells to acquire migratory and invasive capabilities. So, we examined the expression of EMT biomarkers such as E-cadherin, N-cadherin and Vimentin. The results revealed that the expression of epithelial markers such as E-cadherin increased upon stimulation with An-I, whereas the expression of mesenchymal markers such as N-cadherin decreased (Figure 5G and H). However, there is no visible change in Vimentin expression after An-I and GANT61 treatment (Figure 5G and H). In conclusion, An-I decreases the migratory and invasive ability of ovarian cancer cells.

Figure 5:

An-I inhibited migration and invasion of ovarian cancer cells. (A and F) Transwell assay showed that both An-I and GANT61 significantly suppressed migration and invasion of SKOV3 and A2870 cells. (G and H) Western blotting revealed that both An-I and GANT61 remarkably inhibited the EMT of SKOV3 and A2870 cells. **p<0.01; ***p<0.001 (unpaired Student’s t-test). Each experiment was repeated at least three times.

An-I suppresses ovarian cancer tumorigenesis and progression via inhibition of SHH and WNT signaling pathway

In an effort to uncover the mechanisms of antitumor and antimetastatic activity of An-I in ovarian cancer cells, RNA-seq was performed. As expected, GLI1, PTCH1, SNAI1, HHIP and other SHH target genes were upregulated in both SKOV3 and A2870 cells lines compared with IOSE80 and were downregulated by An-I treatment (Figure 6A and B). Interestingly, heat map analysis and gene score enrichment analysis (GSEA) of the differentially expressed genes revealed that the transcripts involved in the WNT pathway were increased in ovarian cancer cells compared with normal ovarian epithelial cells IOSE80 and dramatically decreased following An-I treatment (Figure 6B and C). Moreover, as cancer biomarkers, Matrix metalloproteinases (MMPs) showed the same tendency in ovarian cancer cells treated with An-I (Figure 6A and B). According to previous studies, the SHH transcription repressor, GLI3 repressor (GLI3R), can block WNT signaling and inhibit the transcription of its downstream MMPs by binding to beta-catenin [22, 23]. Therefore, it is reasonable to speculate that An-I downregulates WNT-induced MMPs by controlling SHH signaling activity in ovarian cancer cells, which requires further verification and investigation.

Figure 6:

An-I suppresses ovarian cancer progression via inhibition of SHH and WNT signaling pathway. RNA sequencing was performed on IOSE80, SKOV3, and A2870 cells after treatment with or without An-I. (A) Volcano plot revealed gene expression changes in different samples. (B) GSEA results of An-I-treated and non-treated SKOV3 and IOSE80 were analyzed. (C) A heat map of differentially expressed genes were created.