Atractylenolide III predisposes miR-195-5p/FGFR1 signaling axis to exert tumor-suppressive functions in liver cancer

Background: Antineoplastic activity of atractylenolide III (ATL) has been reported in several malignant tumors. However, its activity has not been completely claried in hepatocellular carcinoma (HCC). Herein, anti-cancer effects and underlying molecular mechanisms of ATL were investigated in HCC cells in vitro. Methods: Cell viability was evaluated by CCK-8 assay. Cell migration and invasion were evaluated using the transwell assay. TUNEL staining was performed to evaluate cell apoptosis. Protein expression was measured by western blotting analysis. On-line database TargetScan and luciferase reporter gene analysis were performed to validate FGFR1 as a target of miR-195-5p. Results: HepG2 and SMMC7721 cell growth, migration and invasion were inhibited by ATL treatment in a dose-dependent pattern. ATL treatment induced apoptosis of HepG2 and SMMC7721 cells. Intriguingly, ATL treatment unexpectedly inhibited FGFR1 protein expression in HepG2 and SMMC7721 cells. Knockdown of FGFR1 inhibited proliferation, migration and invasion, and evoked apoptosis of HepG2 and SMMC7721 cells. We also found that ATL treatment could increase the expression of miR-195-5p, which as a post-transcriptional targeted FGFR1. In HCC tissues, miR-195-5p expression is negatively correlated with FGFR1. Furthermore, the anti-proliferative and pro-apoptotic roles of miR-195-5p were neutralized by overexpressed FGFR1 in HCC cells. Conclusion: ATL effectively repressed growth and induced apoptosis of human HCC cells through the up-regulation of miR-195-5p to down-regulate FGFR1 expression.


Introduction
Hepatocellular carcinoma (HCC) is the most common cause of global cancer-associated mortality with a 1-year survival rate of less than 50% [1]. Patients with advanced HCC have a poor prognosis, meanwhile recurrence and metastasis are often detected even after curative surgery [2]. Currently, palliative chemotherapy, including doxorubicin, mitomycin, cisplatin and 5-uorouracil, is the main method for treatment and improves survival among patients with HCC [3][4][5]. However, nearly all patients develop into resistance to chemotherapeutic agents [6,7]. Thus, new treatment strategies are signi cant for improving therapeutic effects for patients with HCC.
Atractylenolide III (ATL) is a sesquiterpenoid and is the major active ingredient of Atractylodes macrocephala Koidz [8], and is also found in a range of other medical herbs, such as Codonopsis pilosula, Chloranthus henryi Hemsl and Atractylodes lancea. ATL is known to possess anti-in ammatory and anti-cancer properties [9,10]. Recently, ATL promotes human lung carcinoma A549 cells apoptosis by targeting the mitochondria-mediated death signaling pathway [9]. Hence, ATL is believed to be promising for the treatment of GC.
The molecular mechanisms evoked in response to ATL have been investigated in cell culture and animal models [8,11]. Increasing data have been indicated that broblast growth factor receptor1 (FGFR1)/extracellular signal-regulated kinase (ERK) signaling cascade has been involved in the carcinogenesis and maintenance of common cancers [12][13][14]. Elevated levels of FGFR1 have been observed in various types of human cancer, including breast cancer [15], non-small cell lung cancer [16] and gastric cancer [17]. The knockdown of FGFR1 inhibits the growth of MKN-45 cells [18]. Several groups have reported that FGFR1/ERK is highly expressed in HCC, suggesting that FGFR1/ERK may serve as a potential therapeutic target for cancer therapy [12,19,20]. Based on FGFR1 and ERK critical roles in cancer progression, we hypothesized that FGFR1/ERK signaling cascades may be involved in anti-tumor mechanisms of ATL in HCC cell lines. microRNAs (miRs) as non-coding RNAs contributes to post-transcriptional repression that is involved in a variety of physiological and pathological processes, including carcinogenesis [21]. Numerous miRNAs as the tumor suppressor or oncogenes have been reported in hepatic tumorigenesis [22]. A comprehensive analysis of 26 published datasets reveals that a miRNA integrated-signature of 13 miRNAs is differentially expressed in HCC tissues [22]. Among these miRNAs, miR-195-5p as a post-transcriptional regulator of FGFR1 is selected out for molecular mechanism study. miR-195-5p as a tumor suppressor is validated in several cancer subtypes, including prostate cancer [23], bladder cancer [24] and breast cancer [25]. In the present study, we aimed to investigate whether ATL induced growth inhibition and apoptosis in HCC cells by targeting miR-195-5p/FGFR1 signaling pathway.

Sample selection
Human HCC tissues and adjacent para-carcinoma tissues were collected from fty-seven HCC patients undergoing surgical operations at the Xiangya Hospital of Central South University. The clinical research was approved by the Ethics Committee of the Xiangya Hospital of Central South University. Informed consent forms were approved by all of the patients.
Cell culture and drugs.
Human HCC cell lines (MHCC97H, HepG2, Huh7 and SMMC7721) were purchased from the Institute of Biochemistry and Cell Biology of the Chinese Academy of Sciences (Shanghai, China). Cells were maintained in DMEM with 10% fetal bovine serum, 100 U/ml penicillin and 100 μg/ml streptomycin in a humidi ed incubator (Thermo, USA), 5% CO 2 , 95% air atmosphere. ATL was purchased from Sigma-Aldrich (cat. no. A2987; Merck Millipore, Darmstadt, Germany). The molecular structure of Atractylenolide III was shown in Figure 1A.
Cell transfection and plasmid constructs.
Small interfering RNA (siRNA) was constructed to speci cally silence FGFR1. siRNAs were synthesized by GenePharma (Shanghai, China). Target site in human gene encoding FGFR1 was as following: siRNA- Luciferase reporter assay. FGFR1 wild-type (WT) or mutant-type (Mut) 3'-UTR was inserted into the multiple cloning sites of the luciferase-expressing pMIR-REPORT vector (Ambion). For the luciferase assay, HepG2 and SMMC7721 cells containing the WT or Mut 3'-UTR of FGFR1 (0.5 μg) were co-transfected with miR-Con or miR-195-5p mimics using Lipofectamine 2000. The luciferase activity was measured using a luciferase reporter assay kit (Promega Corporation, Madison, WI, USA).
Migration and invasion assays.

RT-qPCR.
Total RNA was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scienti c, Inc., Waltham, MA, USA). TaqMan ® RT kit (Applied Biosystems; Thermo Fisher Scienti c, Inc.) and TaqMan ® MicroRNA assay (Applied Biosystems; Thermo Fisher Scienti c, Inc.) were utilized to detect the expression of miRNAs, according to the manufacturer's protocol. U6 small nuclear RNA was used as an endogenous control.

Immunohistochemical (IHC) staining
The experimental procedures of IHC staining and the scoring rules of FGFR1-positive staining in HCC tissues and adjacent para-carcinoma tissues were referred to as previously described [12]. Primary antibody of FGFR1 (cat. no. ab824; dilution, 1: 100) was obtained from Abcam (Cambridge, UK).

Statistical analysis.
Data from these experiments were reported as mean ± standard deviation for each group. All statistical analyses were performed using PRISM version 7.0 (GraphPad Software, Inc., La Jolla, CA, USA). The student t-test was used to analyze two-group differences. Inter-group differences were analyzed by oneway ANOVA. Differences with a P value of < 0.05 were considered statistically signi cant.

ATL inhibited proliferation in HCC cells.
To analyze the anti-proliferative activity of ATL on HCC cell lines (MHCC97H, HepG2, Huh7 and SMMC7721), cells were exposed to various concentrations of ATLfor 24, 48 and 72 hours, and the CCK-8 assay was used to monitor the cell viability. The results demonstrated that the cell viability of MHCC97H ( Figure 1B), HepG2 ( Figure 1C), Huh7 ( Figure 1D) and SMMC7721 ( Figure 1E) were inhibited by ATL treatment in a dose-dependent manner. We noticed that the cell viability of HepG2 ( Figure 1C) and SMMC7721 ( Figure 1E) was also inhibited by ATL treatment in a time-dependent pattern.
ATL inhibited migration and induced apoptosis in HCC cell lines.
The transwell assay was conducted to assess the migratory and migratory capability of HepG2 and SMMC7721 cells in the presence of ATL with different concentrations for 24 hours. As shown in Figure  2A and Figure 2B, ATL had the ability to restrain migration and invasion in a dose-dependent manner in both the HepG2 and SMMC7721 cells. To study the anti-tumor activity of ATL, cell apoptosis was tested in ATL-treated HepG2 and SMMC7721 cells, which were stained with TUNEL, and then the proportion of TUNEL positive cells was measured by uorescence microscope. As shown in Figure 2C, ATL induced cells apoptosis in a dose-dependent manner in both the HepG2 and SMMC7721 cells.
ATL suppressed FGFR1 in HCC cells.
To examine whether ATL-induced growth inhibition and apoptosis in HepG2 and SMMC7721 cells were due to regulate FGFR1 expression, the protein level was analyzed by western blotting. Our results demonstrated that FGFR1 was dramatically decreased in ATL-treated HepG2 and SMMC7721 cells in a dose-dependent manner ( Figure 3A). To investigate its function in vitro, we constructed small interfering RNA (siRNA) vector targeting FGFR1, namely si-FGFR1. To assess the potential effects of siRNA-mediated FGFR1 silencing on cell proliferation, CCK-8 assay was performed at 24, 48 and 72 hours after si-FGFR1 or siRNA control (si-NC) transfection. Compared with the si-NC group, a signi cant decrease in cell viability was detected in HepG2 and SMMC7721 cells after transfected with si-FGFR1 ( Figure 3A). We further analyzed apoptosis using ow cytometry in HepG2 and SMMC7721 cells after transfected with si-FGFR1 or si-NC ( Figure 3B). Compared with the si-NC group, si-FGFR1 transfected induced cells apoptosis in both the HepG2 and SMMC7721 cells for 48 hours after transfection ( Figure 3C). Furthermore, we analyzed migration and invasion in HepG2 and SMMC7721 cells after transfected with si-FGFR1 or si-NC for 24 hours. The migration and invasion were signi cantly inhibited in HepG2 and SMMC7721 cells after transfected with si-FGFR1 ( Figure 3D and 3E). Thus, our present results indicate that FGFR1 may facilitate HCC malignant progression, inhibition of FGFR1 can inhibit proliferation, migration and invasion, as well as induce apoptosis in HCC cells.
As shown in Figure 5A, miR-195-5p expression was signi cantly decreased in HCC tissues compared with corresponding para-carcinoma tissues. HCC patients with high miR-195-5p expression showed a longer overall survival than those of patients with low miR-195-5p expression ( Figure 5B). IHC staining indicated that FGFR1 protein expression exhibited a signi cant elevation in HCC tissues compared with corresponding para-carcinoma tissues ( Figure 5C and 5D). HCC patients with high FGFR1 expression showed a shorter overall survival than those of patients with low FGFR1 expression ( Figure 5E). Correlation analysis suggested that miR-195-5p expression is negatively correlated with FGFR1 in HCC tissues ( Figure 5F). miR-195-5p-caused growth inhibition and apoptosis were reversed by overexpressed FGFR1. miR-195-5p mimics and FGFR1 overexpressed plasmids were co-transfected into HepG2 and SMMC7721 cells. Overexpression of miR-195-5p signi cantly suppressed proliferation ( Figure 6A) and induced apoptosis ( Figure 6B and 6C), as well as blocked migration and invasion ( Figure 6D) in HepG2 and SMMC7721 cells. However, the tumor-suppressive functions of miR-195-5p were neutralized by overexpressed FGFR1 in HepG2 and SMMC7721 cells ( Figure 6A-6D). These ndings suggest that miR-195-5p and FGFR1 may play reciprocal roles in the carcinogenesis of HCC. These results suggest that ATL might be a potential therapeutic drug for HCC, and the underlying mechanism was mediated, at least partially, through the up-regulation of miR-195-5p to repress FGFR1 expression (Figure 7).

Discussion
The present study investigated the anti-proliferative and pro-apoptotic effects of ATL in HCC cells. When ATL was exposed to the HepG2 and SMMC7721 cells, the growth, migration and invasion of these cell lines were dramatically inhibited. Moreover, ATL treatment resulted in a signi cant increase in cell apoptosis of HCC cells. Intriguingly, we observed that ATL unexpectedly inhibited FGFR1 in HCC cells. The knockdown of FGFR1 also inhibited proliferation and induced apoptosis in HCC cells. In addition, we discovered that ATL treatment led to the up-regulation of miR-195-5p expression, which as a posttranscriptional targeted FGFR1. The anti-proliferative and pro-apoptotic roles of miR-195-5p were neutralized by overexpressed FGFR1 in HCC cells.
The activation of the apoptotic pathway in cancer cells can serve as a defensive mechanism against the proliferation of cancer cells [26]. Pro-apoptotic protein, BAX, and anti-apoptotic protein, Bcl-2, are closely related to mitochondrial apoptosis by enhancing mitochondrial membrane permeabilization and activating caspase-3 expression [27]. First, we detected the effects of ATL on cell apoptosis by TUNEL staining, which showed that the percentage of apoptotic cells was signi cantly increased and positively related to the concentrations of ATL in HCC cells. Taken together, these ndings indicate that ATL induces apoptosis in human HCC cells and may be a potential drug for cancer therapy.
Mounting evidence has shown that FGFR signaling plays crucial roles in cancer cell proliferation, migration, angiogenesis and survival by targeting mitogen-activated protein kinase (MAPK) signaling pathway, including ERK, p38 MAPK and c-Jun N-terminal kinase (JNK) pathways [12,28]. Fibroblast growth receptor family includes FGFR1, 2, 3 and 4 that serve as receptor tyrosine kinases [29]. FGFR1 activation has been found in a number of human cancers and plays a crucial role in drug resistance, cancer progression and metastasis [19,30,31]. Wang et al have been suggested that FGFR1 can be regulated by non-coding RNA and can serve as a potential therapeutic target for HCC [12]. To explain the possible mechanism underlying ATL-induced HCC cells apoptosis, we examined the protein expression levels of FGFR1 in HCC cells. Strikingly, FGFR1 was signi cantly inhibited in ATL-treated HCC cells.
Inhibition of FGFR1 showed a signi cant growth inhibition and apoptosis in HCC cells. Our results suggest that ALT may possess the anti-cancer capability that provides a new potential therapeutic drug for HCC treatment.
We also discovered that ALT treatment increased the expression of miR-195-5p in HepG2 and SMMC7721 cells. Based on this conclusion, we attempted to explore the roles of miR-195-5p in the progression of HCC. In vitro experiments validated that over-expression of miR-195-5p inhibited proliferation and promoted apoptosis in HCC cells, as well as reduced the protein expression of FGFR1. Interestingly, the antineoplastic activities of miR-195-5p were counteracted by overexpressed FGFR1 in vitro. Shi et al demonstrate that the expression of miR-195-5p is decreased in the HCC tissues, which was also validated on the TCGA dataset [22]. Zhu et al also exhibit that miR-195-5p is lower in the HCC tissues than in the matched non-tumor tissues [32]. Xu et al indicate that miR-195-5p acts as an anti-oncogene in HCC, re ecting that overexpression of miR-195-5p leads to suppression of invasion, migration and proliferation in vitro [33]. Consistent with these results, we also found that overexpression of miR-195-5p showed an anti-proliferative and pro-apoptotic activities by inhibiting FGFR1 expression in HCC cells.
In summary, our study has shown the potential of ATL for the e cient induction of human HCC cells growth inhibition and apoptosis, and the underlying mechanism is mediated, at least partially, via targeting miR-195-5p/FGFR1 signaling pathway. Consequently, we will further study the chemosensitization strategies of ATL in vivo to gain insights to aid in the development of an effective strategy for reversing the drug resistance of HCC.

Declarations
Ethics approval and consent to participate: The clinical research was approved by the Ethics Committee of the Xiangya Hospital of Central South University. Informed consent forms were approved by all of the patients.

Consent for publication:
Not applicable.

Availability of data and material:
The data that support the ndings of this study are available from the corresponding author upon reasonable request.  ATL inhibited migration and invasion in HCC cell lines. The transwell assay was conducted to assess migration and invasion of HepG2 (A) and SMMC7721 (B) cells in the presence of ATL with different concentrations for 24 hours. Cell apoptosis was evaluated using TUNEL staining in HepG2 and SMMC7721 cells with ATL treatment for 48 hours (C). n= 3 in each group. * P < 0.05, ** P < 0.01 and *** P < 0.001 compared with control group. ATL suppressed FGFR1 in HCC cells. HepG2 and SMMC7721 cells were exposed to ATL with various concentrations for 48 hours, the protein expression of FGFR1 was measured by western blotting assay (A). After transfected with siRNA-FGFR1 for 48 hours, cell viability was detected by CCK-8 assay (B); cell apoptosis was analyzed using TUNEL staining (C); the transwell assay was conducted to assess migration and invasion in HepG2 and SMMC7721 cells (D and E). n= 3 in each group. * P < 0.05, ** P < 0.01 and *** P < 0.001 compared with control group.  Complementary pairing sequences between miR-195-5p and FGFR1 were predicted using TargetScan (B). Luciferase reporter assay was performed to verify FGFR1 as a direct target of FGFR1 (C). After transfection with miR-Con or miR-195-5p mimics into HepG2 and SMMC7721 cells, the protein expression of FGFR1 was detected using western blotting (D). * P < 0.05, ** P < 0.01 and *** P < 0.001 compared with corresponding control group. Figure 5 miR-195-5p expression was negatively correlated with FGFR1 in HCC tissues. miR-195-5p expression in fth-seven pairs of HCC tissues and corresponding para-carcinoma tissues was measured using RT-qPCR (A). HCC patients with high miR-195-5p expression showed a longer overall survival than those of patients with low miR-195-5p expression (B). IHC staining was used to evaluated FGFR1 protein in HCC tissues and corresponding para-carcinoma tissues (C and D). HCC patients with high FGFR1 expression showed a shorter overall survival than those of patients with low FGFR1 expression (E). Pearson correlation analysis was used to analyze the correlation between miR-195-5p and FGFR1 in HCC tissues (F). Figure 6 miR-195-5p-caused growth inhibition and apoptosis were reversed by overexpressed FGFR1. After cotransfection with miR-195-5p mimics and FGFR1 overexpressed plasmids into HepG2 and SMMC7721 cells, cell viability was detected by CCK-8 assay (A); cell apoptosis was measured using TUNEL staining (B and C); the transwell assay was conducted to assess migration and invasion (D). * P < 0.05 compared with corresponding control group; # P < 0.05 compared with mimics + vector-Con group. A schematic diagram for the effect of atractylenolide III to exert tumor-suppressive functions in liver cancer by targeting miR-195-5p/FGFR1 signaling axis.