Liensinine Inhibits Osteosarcoma Growth by ROS-Mediated Suppression of the JAK2/STAT3 Signaling Pathway

Osteosarcoma (OS) is the most common malignancy of bone. Liensinine exerts antitumor effects on cancers of the colon, breast, and gallbladder. However, its antitumor activity in OS remains unclear. This study is aimed at investigating the efficacy of liensinine against OS and the underlying mechanism of action. Cell proliferation, apoptosis, and cycle arrest in OS were detected using the Cell Counting Kit-8 (CCK-8), colony formation, and flow cytometry assays, respectively. The production of reactive oxygen species (ROS), glutathione (GSH) and glutathione disulfide (GSSG) concentrations, and mitochondrial membrane potential (MMP) of OS cells were measured by flow cytometry, colorimetry, and JC-1 staining. The expressions of factors related to apoptosis, cell cycle, and activation of the JAK2/STAT3 pathway were determined by Western blotting. To examine the potential role of ROS, an antioxidant (N-acetyl cysteine, NAC) was used in combination with liensinine. In vivo, we generated a xenograft mouse model to assess its antitumor efficacy. Tissue level expressions of factors related to apoptosis and activation of the JAK2/STAT3 pathway were assessed by immunohistochemistry or Western blotting. Liensinine inhibited the proliferation and induced G0/G1 phase arrest and apoptosis of OS cells in a dose-dependent manner. Additionally, liensinine promoted intracellular ROS production, enhanced the GSSG/GSH ratio, and induced MMP loss and ROS-mediated suppression of the JAK2/STAT3 pathway. NAC significantly attenuated the liensinine-induced antitumor activities and activated the JAK2/STAT3 pathway. In vivo, liensinine effectively inhibited the OS growth and promoted apoptosis; however, it had no negative effect on the internal organs. In conclusion, liensinine-induced ROS production could suppress the activation of the JAK2/STAT3 pathway and inhibit the OS growth both in vivo and in vitro. Our findings provided a new rationale for subsequent academic and clinical research on OS treatment.


Introduction
Osteosarcoma (OS) is a malignant tumor originating from mesenchymal tissues, accounts for 20% of the primary malignant bone acanthomas; it is the most common primary malignant bone tumor in adolescents [1]. OS is characterized by high malignancy, rapid growth, early metastasis, and poor prognosis. It is the leading cause of cancerrelated death among children and adolescents [2,3]. At present, surgery combined with multiagent chemotherapy is the standard treatment regime for OS [2,4]. However, longterm chemotherapy causes many irreversible systemic side effects, including cardiotoxicity, secondary malignancies, neurotoxicity, and infertility. In addition, physical disability caused by surgical procedures has a great impact on the mental health and quality of life of the patients [2,5]. Therefore, the development of novel anti-OS drugs is important for effectively inhibiting tumor progression and prolonging the survival duration of patients.
Liensinine, a natural isoquinoline alkaloid, is isolated from the seed embryo of Nelumbo nucifera Gaertn [6]. Multiple biological effects of liensinine are reported in recent pharmacological studies, including antihypertension and antiarrhythmias, as well as anticancer properties [6][7][8]. For example, liensinine can induce apoptosis and growth cessation of gallbladder cancer [9]. It can also induce apoptosis of colorectal cancer by stimulating mitochondrial dysfunction and sensitize breast cancer cells towards chemotherapy through DNM1L-mediated mitochondrial fission [6,10]. However, unlike the aforementioned cancer types, OS is distinct in both origin and biological behavior. The role of liensinine in OS remains unknown and warrants investigation.
Reactive oxygen species (ROS) are the by-products of oxidative stress, which include peroxides, superoxides, hydroxyl radicals, and singlet oxygen [11]. Accumulating evidence demonstrates that several anticancer drugs, such as Glaucocalyxin A, metformin, cisplatin, isoalantolactone, and docetaxel, can induce apoptosis and proliferation cessation of cancer cells by intracellular ROS generation [11][12][13]. According to previous studies, the inhibitory effect of liensinine on gastric cancer cells is achieved by inducing ROS production [14]. In addition, other similar dibenzyl tetrahydroisoquinoline alkaloids, such as cepharanthine or tetrandrine, can also induce ROS-mediated apoptosis of tumor cells [15,16]. Typically, tumor cells can withstand oxidative stress by maintaining the balance between oxidation and antioxidation levels; however, antineoplastic drugs can fatally break the redox balance not just by increasing oxidant accumulation but also by interfering with the antioxidant production [11]. The role of liensinine in the oxidation/antioxidation system of OS cells is unknown.
In this study, we aimed to investigate the antitumor effect of liensinine on OS and further explore its underlying mechanism of action, so as to elucidate the regulation of ROS on the JAK2/STAT3 pathway.

Materials and Methods
2.1. Cell Culture. The source and culture conditions of human normal osteoblasts, hFOB 1. 19, and human OS cell lines SaOS-2, MG-63, 143B, and U-2OS were the same as described in our previous study [17].

Colony Formation
Assay. 1000 cells per well were plated onto 6-well plates (each well contained 2 mL medium supplemented with 10% FBS). After adherence for 24 h, the cells were treated with 0 μM, 40 μM, and 80 μM liensinine for 24 h; following which, the medium was replaced with the complete medium. The cells were cultured for 14 days; subsequently, they were fixed with 4% ice-cold paraformaldehyde, stained with 0.1% crystal violet, and washed thrice with PBS (P1010, Solarbio, Beijing, China). Images of cell colonies were captured by a camera (Alpha 7R IV, SONY, Tokyo, Japan), and the cell colony numbers were analyzed using the Image J software.
2.5. Flow Cytometric Analysis of Apoptosis. The apoptosis was evaluated by flow cytometry (FCM) using an Annexin V-FITC/PI Apoptosis Detection Kit (C1062L, Beyotime). After SaOS-2 and 143B cells were treated with liensinine or cotreated with NAC and liensinine in 6-well plates for 24 h, approx. 1 × 10 6 cells/well were harvested and detected as described previously [19].
2.6. Flow Cytometric Analysis of Cell Cycle Distribution. The cell cycle distribution of cells was evaluated by FCM using a Cell Cycle Detection Kit (C1052, Beyotime). After SaOS-2 and 143B cells were treated with liensinine for 24 h, approx. 1 × 10 6 cells/well were harvested and washed twice with PBS. The resuspended cells were then fixed with prechilled 70% ethanol (E111994, Aladdin, Shanghai, China) for 2 h at 4°C. Following this, the collected cells were incubated with 500 μL PI/RNase A staining fluid for 30 min at 37°C in dark. The cell cycle distribution was detected and analyzed on a CytoFLEX flow cytometer (Beckman Coulter, Brea, CA, USA) with 488 nm excited wavelength.

Flow
Cytometric Analysis of ROS. The ROS production in the cells was evaluated by FCM using a DCFH-DA probe-based ROS Detection Kit (S0033, Beyotime). DCFH-DA was diluted with FBS-free medium to a final concentration of 10 μM. After SaOS-2, 143B, and hFOB 1.19 cells were treated with liensinine or cotreated with NAC and liensinine for 24 h, the cells (1 × 10 6 ) were harvested and washed twice with PBS. The cells were then incubated with 500 μL of 10 μM DCFH-DA for 20 min in a 37°C incubator and washed thrice with FBS-free medium. Finally, the ROS in the cells was detected and analyzed on a CytoFLEX flow cytometer with 488 nm excited wavelength.

Colorimetric Analysis of Glutathione (GSH) and
Glutathione Disulfide (GSSG). The level of GSH and GSSG in the cells was evaluated by colorimetry using a GSH and GSSG Detection Kit (S0053, Beyotime). After SaOS-2 and 143B cells were treated with liensinine or NAC, the cells were collected and washed with PBS. 30 μL of protein removal reagent M solution was added. The samples were then subjected to two rapid freeze-thaws using liquid nitrogen and water at 37°C. After the samples were placed at 4°C for 5 min, they were centrifuged for 10 min at 10,000 × g, and the supernatant was collected for the determination of total glutathione. In addition, part of the above supernatant was pipetted and added with the GSH clearance auxiliary solution as well as the GSH clearance working solution at the ratios of 100 μL : 20 μL and 100 μL : 4 μL. The solutions were mixed by vortex and reacted for 60 min at room 2 Oxidative Medicine and Cellular Longevity temperature to obtain samples for GSSG determination. In a 96-well plate, 10 μL of the supernatant was mixed with 150 μL of the prepared total glutathione detection working solution and incubated at room temperature for 5 min, followed by the addition of 50 μL of 0.5 mg/mL NADPH. After 25 minutes of reaction, the absorbance of each well was measured by a microplate reader at 412 nm wavelength. GSSG was determined in the same way. The GSH and GSSG levels were further calculated according to the kit manufacturer's protocol and the formula: GSH = total glutathione − GSSG × 2.
2.9. JC-1 Staining. The mitochondrial membrane potential (MMP) of the cells was evaluated by a JC-1 Kit (C2006, Beyotime). After SaOS-2 and 143B cells were treated with liensinine or NAC in the 6-well plates for 24 h, the medium was discarded and 1 mL JC-1 working solution was added to the wells, and cells were incubated for 20 min at 37°C. The cells were then washed twice with 1× JC-1 staining buffer. After adding 1 mL PBS per well, the fluorescence signal of the cells was observed under an inverted fluorescence microscope (DMI8, Leica) at ×200 magnification. Polymers were excited by green light, monomers were excited by blue light, and merged images were collected.
2.10. In Vivo Assay. The animal study design was approved by the ethics committee of Peking University Third Hospital (No. IRB00006761-2016048) and complied with the ARRIVE guidelines. All experiments with animals were performed in the Department of Laboratory Animal Science, Peking University Health Science Center. Eighteen 4-weekold BALB/c nude female mice weighing 18 ± 2 g were purchased from Beijing WeiTongLiHua Laboratory Animal Technology Co., Ltd. The mice were kept in a specific pathogen free (SPF) environment and provided an ad libitum diet. After feeding for five days, the mice adapted to the environment; next, 143B cells at a density of 2 × 10 6 /100 μL were subcutaneously injected into the right axilla of each mouse. When the tumor volume reached 100 mm 3 , the mice were randomly divided into three groups: (1) sham treatment (n = 6), (2) liensinine-L treatment (n = 6), and (3) liensinine-H treatment (n = 6). The mice in the three groups were orally administered with 200 μL PBS, 15 mg/kg liensinine solution, or 30 mg/kg liensinine solution, respectively, once every two days for 14 days. The tumor sizes and body weights were measured every two days with the volume calculation formula: L × W2/2 (L, the long diameter of the tumor and W, the short diameter of the tumor). On the last day, the mice were sacrificed. Subcutaneous tumors from each mouse were excised and weighed; one portion was immediately frozen for Western blot analysis, while another was preserved in 4% paraformaldehyde for histopathological analysis. In addition, the heart, liver, spleen, lung, and kidney from each mouse were removed, photographed, and collected for later use.

Oxidative Medicine and Cellular Longevity
About 200 μL serum was collected for liver and renal function detection in the above institution.
2.14. Statistical Analysis. Statistical significances for independent data were assessed using an unpaired Student's t -test. For multiple comparisons, one-way ANOVA was used to analyze the data with post hoc Bonferroni correction. The data were analyzed using the IBM SPSS Statistics 22.0 software (IBM, Armonk, NY, USA), and the results were depicted as mean ± standard deviation ðSDÞ. Differences were considered significant at P < 0:05.

10
Oxidative Medicine and Cellular Longevity p-STAT3 and p-JAK2, thereby reducing the ratio of p-STAT3/STAT3 and p-JAK2/JAK2. These results were negatively correlated with Bax expression and positively correlated with that of Bcl-2. Therefore, we reasonably speculated that liensinine-induced oxidative stress suppressed the activation of the JAK2/STAT3 pathway and further inhibited biological activity of OS cells. As the control, we also evaluated the effect of the same concentration of liensinine on apoptosis (cleaved caspase 3 and cleaved PARP1) and JAK2/STAT3 pathway activation (p-STAT3, STAT3, p-JAK2, and JAK2) of hFOB 1.19 cells. The results showed that the expressions of the above proteins remained unchanged upon liensinine treatment, which implied that, liensinine did not induce excessive apoptosis and activation of the JAK2/STAT3 pathway in human normal osteoblasts ( Figures S1(b) and S1(c)).  (Figure 7(e)). In addition, H&E staining results of the heart, liver, spleen, lung, and kidney excised from the mice showed that liensinine had no negative effect on internal organs as compared to the control group (Figure 7(f)). Through Western blot analysis of tumor tissues, we found that liensinine treatment significantly downregulated the expressions of p-STAT3 and p-JAK2, thereby causing a significant reduction in the ratio of p-STAT3/STAT3 and p-JAK2/JAK2 (Figures 7(f) and 7(g)). These results were consistent with the suppression of JAK2/STAT3 pathway by liensinine in vitro. In addition, no significant change was found in the routine blood test as well as the liver and renal function blood test (Figures 7(i) and 7(j); Figure S1(d)-S1(g)). Thus, the administration of liensinine was biocompatible in mice.

Discussion
Traditional Chinese medicine (TCM) has been widely used to treat various diseases for thousands of years. Among them, liensinine, a natural isoquinoline alkaloid, can be isolated from a Chinese medicinal herb (Nelumbo nucifera Gaertn) and possesses several medicinal benefits [6][7][8].
Increasing evidence shows that liensinine may be a potential antitumor agent, and its efficacy against human colorectal, gallbladder, and breast cancers has been validated by 14 Oxidative Medicine and Cellular Longevity previous studies [6,9,10]. As a malignant tumor originating from mesenchymal tissue, OS has faster progression, earlier hematogenous metastasis, and fewer targeted drugs available than the aforementioned cancer types. Thus, it requires more radical resection as it causes more misery to the patients [22]. Therefore, it is important to develop efficient and safe pharmacotherapy for OS. To date, the effect of liensinine in OS has not been reported, both in vitro and in vivo.
In this study, liensinine was found to significantly reduce the viabilities and cloning efficiencies of multiple OS cell lines in a dose-dependent manner; it exerted no negative effect on human normal osteoblasts, even at a relatively higher dose. Tumor cells usually have higher metabolic activities and ROS levels as compared to the normal cells and are more vulnerable to liensinine-induced ROS insults [17,23]. The mechanism underlying liensinine-mediated elevation in oxidative stress in tumor cells requires further study. In addition, tumor cells show higher cellular uptake of natural compounds but have lower intracellular glutathione levels than normal cells, which may be related to the selective cytotoxic effect of liensinine on OS cells [24,25]. The results of flow cytometry analysis indicated that liensinine could substantially block G0/G1 progression into S phase and promote apoptosis in OS cells, consistent with the previous studies that reported the apoptosis-induction effect of liensinine on other cancer cell types [9,10,14]. We also confirmed the antitumor activity of liensinine in xenografted-tumor mice, which further validated our in vitro findings as indicated by the subcutaneous tumor sizes and immunohistochemical results post 2-week treatment. Moreover, according to the H&E staining of the visceral slices, routine blood test, and liver and renal function test, we concluded that the appropriate dosage of liensinine was safe and reliable for in vivo application.
To validate the above findings at the molecular level, we performed Western blot analysis. The imbalance in the expressions of Bax and Bcl-2 is involved in mitochondrial apoptosis, inducing the expression of cleaved caspase 3 and promoting tumor apoptosis [26]. Once caspase 3 is cleaved and activated, cascading effects ensue, and the apoptosis process is irreversible. Therefore, caspase 3 is considered to be a key factor in triggering apoptosis [27]. PARP is a nuclear protein that is activated by DNA damage and participates in DNA repair. However, it is also a substrate of the caspases, which can lose enzymatic activity and accelerate apoptosis after cleavage [28,29]. Cyclin D1 is a regulatory subunit of cyclin-dependent kinases (CDKs) that promotes G0/G1 progression [30]. Hence, changes in expressions of Bcl-2, Bax, cleaved caspase 3, cleaved PARP, and Cyclin D1 reflect the apoptosis and cell cycle status. After liensinine treatment, Bax, cleaved caspase 3, and cleaved PARP were found to be upregulated, while those of Bcl-2 and Cyclin D1 were downregulated in OS cells, which indicated that liensinine may be a potential agent for inhibiting OS progression.
The key roles of ROS in biological processes include cell signaling, biosynthetic processes, and host defense [31]. A  Representative items: ALT: alanine aminotransferase; Cr: creatinine. n = 6 in each group of (b-d, i, j). n = 3 in each group of (g). * P < 0:05, * * P < 0:01, and * * * P < 0:001 vs. sham. 17 Oxidative Medicine and Cellular Longevity mild elevation of ROS levels and ROS scavenging rate within tumor cells, termed as "mild oxidative" stress, is their normal physiological state and is associated with the activation of oncogenic pathways [32,33]. However, sustained accumulation of ROS, especially free radicals, will cause damage to DNA, lipids, and proteins, eventually leading to adverse cellular outcomes [11][12][13]. Many drugs exert potent cytotoxic effects through this mechanism, for example, Glaucocalyxin A inhibits OS progression by inducing excessive oxidative stress [11]. GSH is a potent antioxidant that eliminates intracellular ROS through a redox reaction, by itself being oxidized to GSSG [34]. The imbalance of GSSG/GSH system reflects a notable deficiency in ROS scavenging capacity [11,35]. To explore whether liensinine can promote ROS production, the levels of ROS, GSSG, and GSH in OS cells were evaluated. The ROS and GSSG levels were found to be upregulated while that of GSH was downregulated after treatment, suggesting that liensinine could induce excessive oxidative stress in OS cells. With the decrease of MMP, ROS increase and are released from mitochondria into the cytoplasm, further resulting in excessive oxidative stress, which subsequently triggers more release of ROS from the neighboring mitochondria, ultimately leading to cell death [36]. A decline in MMP is therefore a hallmark event in the early stage of apoptosis [37]. We found significant MMP loss in liensinine-treated tumor cells by JC-1 staining, confirming the occurrence of oxidative stress damage. Moreover, pretreatment with NAC, an antioxidant or ROS scavenger, could significantly reduce ROS production and apoptosis of OS cells stimulated by liensinine, which indi-cated that the liensinine-induced oxidative stress injury was directly responsible for the inhibition of OS survival.
Previous studies have demonstrated that the biological activities of tumor cells are regulated by multiple signaling pathways including the JAK2/STAT3 cascade [20,21,38,39]. Activation of the JAK2/STAT3 pathway can promote cell survival and proliferation, inhibit apoptosis, and is implicated in the occurrence and development of various cancers [40,41]. Its antiapoptotic effect may be realized by reducing the Bax/Bcl-2 apoptotic switch ratio and inhibiting the activation of key proapoptotic enzymes in cells [42,43]. In this study, we first reported that liensinine could suppress the activation of the JAK2/STAT3 pathway in tumor cells. Furthermore, excessive intracellular ROS can induce cell cycle arrest and apoptosis [32,44]. Therefore, we reasonably speculated that the liensinine-induced intracellular ROS accumulation may suppress the activation of the JAK2/ STAT3 pathway, thereby inhibiting the proliferation and promoting the apoptosis of OS cells. The mediating role of ROS was confirmed when NAC was found to significantly reverse the liensinine-induced JAK2/STAT3 pathway blockade. Together, the results suggested that liensinine could exert antitumor effect by affecting JAK2/STAT3 pathway activation through excessive oxidative stress (Figure 8). The finding that the JAK2/STAT3 pathway could be blocked by excessive ROS was consistent with the tumor-related reports [11,13,45], but contrary to the conclusions of some inflammation-related studies [46,47]. This may be due to differential roles and mechanisms of ROS in tumor cells and macrophages, which requires further exploration.

18
Oxidative Medicine and Cellular Longevity In conclusion, the present study showed that the liensinine-induced ROS production could suppress the activation of the JAK2/STAT3 pathway and inhibited the OS growth both in vivo and in vitro. Our findings provided a new rationale for subsequent academic and clinical research on OS treatment.