Sanguinarine triggers intrinsic apoptosis to suppress colorectal cancer growth through disassociation between STRAP and MELK

Antibodies against Caspase-3 (No. 9662), PARP (No. 9532), Pro-caspase 9 (No. 9508), Cleaved Caspase-9 (No. 7237), Cytochrome C (No. 4272), Bcl-2 (No. 2870), Bax (No. 5023) and PCNA (No. 13110) were from Cell Signaling Technology (CST), USA. Anti-maternal embryonic leucine zipper kinase (MELK) (No. ab155767) was obtained from Abcam, UK. Anti-serine-threonine kinase receptor-associated protein (STRAP) (No. sc-377,345) and Caspase 8 (No. sc-56,070) antibody was purchased from Santa Cruz Biotechnology, USA. Anti-phospho-MELK (Thr167, Ser171) (No. WG-00203P) and Anti-phospho- STRAP (Thr175, Ser179) (No. WG-00204P) were ordered from ABclonal Technology, China. Alexa fluor^® 488 goat anti-rabbit IgG(H + L) (No. CA11008s) and Alexa fluor^®546 goat anti-mouse IgG(H + L) (No. A11003) were obtained from Molecular Probes (Invitrogen). Beta-actin antibody (No. E021020–01) was from EarthOx, LLC, San Francisco, USA. Sanguinarine was purchased from National Institutes of Food and Drug Control (Beijing, China). Sanguinarine was dissolved in DMSO (MP, France) and diluted in culture medium for each experiment. The final concentration of DMSO didn’t exceed 0.1%.

Human colon carcinoma cell lines SW480 (cat. No. TCHu172) and HCT116 (WT) (cat. No. TCHu 99) were purchased in 2014 from the Chinese Academy of Science Committee Type Culture Collection Cell Bank (Shanghai, China). HCT116 Bax ^−/− human colon carcinoma cell line was a generous gift from Dr. Bert Vogelstein (Howard Hughes Medical Institute, MD). SW480 and HCT116 (WT) cells were cultured in RPMI 1640 medium (Gibco) and HCT116 Bax ^−/− cells in McCoy’s 5A medium (Invitrogen). The medium was supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin-streptomycin (Invitrogen). All cells were maintained at 37 °C in a humidified incubator of 5% CO₂-containing atmosphere.

Cells were seeded in 96 wells plates at density of 0.8 × 10⁴ cells per well. Sanguinarine was dissolved in DMSO as a stock solution and was stored in aliquots at − 20 °C. After treatment with or without sanguinarine, MTT (3-[4, 5-dimethylthiazolyl-2]-2, 5-diphenyl tetrazolium bromide) (MP, France) was added into each well. Cells were incubated for 4 h. The formazan crystals were dissolved in DMSO. The absorbance value was measured at 490 nm.

Cells were stained with PE-Annexin V/7-AAD (BD Bioscience pharmingen) according to the manufacturer’s instructions. Cells were resuspended in 500 μL 1 x binding buffer, 5 μl PE-Annexin V and 5 μl 7-AAD were added to the sample and incubated at room temperature for 15 min in the dark. The stained samples were then detected by flow cytometry (BD FACSCalibur).

Total cells were lysed in RIPA buffer. Mitochondrial and cytoplasmic fractions were isolated according to the manufacturer’s instructions of Cell Mitochondria Isolation Kit (Beyotime, China). Mitochondria were lysed in RIPA buffer. The protein concentrations of the mitochondria, cytoplasm and whole cells were determined by the BCA (Thermo scientific) method using the Thermo protein assay kit. Equal amount of proteins from each group were subjected to SDS-PAGE on 12% gel, transferred to a PVDF membrane (Millipore) by electroblotting, blocked with 5% nonfat milk for 1 h at room temperature and incubated with primary antibodies overnight at 4 °C. Next, the membrane was incubated by HRP-conjugated appropriate secondary antibodies, visualized by enhanced chemiluminescence (Millipore) and exposed by KODAK Image Station 4000MM Digital Imaging System [21, 22].

Cells were stained with JC-1 according to the manufacturer’s instructions of Mitochondrial membrane potential assay kit (Beyotime, China). Cells were then detected using flow cytometry (BD FACSCalibur). The mitochondrial membrane potential was calculated based on the following equation: mitochondrial membrane potential = red fluorescence intensity / green fluorescence intensity.

After treatment with sanguinarine of different concentration for indicated time, cells were stained with 5-(and 6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) according to the manufacturer’s instructions of Reactive Oxygen Species Assay Kit (Beyotime, China). Intracellular production of ROS was evaluated by flow cytometry. Relative Ros levels = DCH fluorescence intensity(treatment group) / DCH fluorescence intensity (control group).

Total RNA was extracted with TRIzol (Takara, Dalian, China). 2 μg of the total RNA was reversely transcribed using a reverse transcription kit (Takara) according to the manufacturer’s protocols and cDNA was obtained. The primers for STRAP were 5’-AAGGGACACTTTGGTCCTATTC-3′ (fwd), 5’-CCTACCACAGTTTGCCATAGT-3′ (rev). The primers for MELK were 5′- ACTTGCCTGCCATATCCTTAC-3′ (fwd), 5’-GGTTCTTCAAGGCCTCAATCT -3′ (rev).The primers for GAPDH were 5’-ATTGTCAGCAATGCATCCTG-3′ (fwd), and 5′-ATGGACTGTGGTCATGAGCC-3′ (rev). All reactions were performed in triplicate for 40 cycles in a Stratagene Mx3005P system as previously described [23]. Relative expression was calculated with GAPDH using the 2^-ΔΔCt.

Immunoprecipitation was performed as previously described [24]. Briefly, cell extract was mixed with anti-STRAP, anti-MELK antibody at 4 °C overnight. Agarose beads were added at a ratio of 1 mg of extract per 120 μl of agarose at 4 °C for 3 h. The beads were then pelleted at 2500×g for 3 min and washed with lysis buffer five times. The beads were subjected to elution with 5 vol of 0.5 mg/ml peptide for 4 h or directly boiled in loading buffer.

Cells were fixing with 4% paraformaldehyde at room temperature for 25 min and permeabilized using 0.2% Triton-100X and then blocked by 5% BSA at room temperature for 1 h. Next, they were incubated with primary antibody overnight at 4 °C according to the manufacturer’s instructions. Then, the cells were followed by staining with secondary antibodies (Alexa Fluor 488 and 546, Invitrogen). Cell nuclei were stained with DAPI at 37 °C in darkness for 8 min. Pictures were taken under confocal microscope (NIKON C2⁺).

The in vitro kinase activity of MELK was analyzed by using HTRF^® KinEASE™-STK kit. It is a generic method for measuring Serine/Treonine kinase activities. The assay was performed according to manufacturer’s protocol. MELK Kinase was incubated with sanguinarine and the STK Substrate-biotin for 50 min at room temperature in the presence of ultrapure ATP. STK Antibody labeled with Eu3⁺-Cryptate STK Substrate-biotin and streptavidin-XL665 were added to the mixture in a single addition with EDTA. They were incubated for 60 min at room temperature to stop the kinase activity. The kinase activity of MELK were detected by Tecan Infinite® 200 PRO microplate spectrophotometer.

6 to 8 week-old BALB/c-nu male nude mice (NO. 44002100007706) were purchased from Experimental Animal Centre of Southern Medical University. The mice were housed in a specific pathogen-free facility and maintained under 12 h light/dark cycles. All animal studies were performed according to protocols of the Care and Use of Laboratory Animals published by the National Institutes of Health and approved by the Laboratory Animals Care and Use Committee of Southern Medical University. Orthotopic CRC model on nude mice was established according to our previously described methods [24]. Twenty-Four tumor-bearing male nude mice were randomly divided into four groups (n = 6 each group) treated with (1) Control diluents; (2) sanguinarine 4 mg/kg/d by administeration via oral gavage; (3) sanguinarine 8 mg/kg/d by administeration via oral gavage; (4) cisplatin 1 mg/kg/d by intraperitoneal injection for 21 days. Nude mice were weighted every 5 days. The mice were anesthetized with Sodium Pentobarbital 100 mg/kg and then sacrificed at the end of the experiment according to the euthanasia guidelines for experimental animals of Southern Medical University. The tumor tissues were dissected and fixed in 4% paraformaldehyde for histologic examination or stored at − 80 °C for western blotting.

The collections of tumor and adjacent tissue samples were approved by the Ethics Committee of Guangdong General Hospital and all aspects of the study comply with the Declaration of Helsinki. Informed consents were signed voluntarily by all patients that included in this study. Fresh primary CRC specimens and paired noncancerous colorectal tissue were provided by the Tumor Tissue Bank of Guangdong General Hospital. In each case, a diagnosis of primary CRC had been made, and the patient had undergone elective surgery for CRC in Guangdong General Hospital between 2015 and 2016. The pathological diagnosis was made in the Department of Pathology of Guangdong General Hospital.

Tumour tissues from control and treated mice were obtained after sacrificing the mice and then fixed in 4% paraformaldehyde and embedded in paraffin. Sections were deparaffinized, rehydrated. Endogenous peroxidase activity was blocked with 3% (v/v) hydrogen peroxide solution. Heat-induced antigen retrieval was performed. Sections were incubated with 5% BSA to block no-specific staining. After incubation with the primary antibodies (anti-MELK, anti-STRAP, anti-Bax) overnight at 4 °C in a humid chamber, sections were incubated with mouse/rabbit-labelled polymer from GTVision™ III kit (Gene Tech, China) for 1 h at 37°C. Positive signals were visualized by DAB kit. The slides were reviewed by two or three pathologists blind to the study. To evaluate the expression levels, immunostained slides were evaluated using a method described previously [25]. TUNEL assay was performed according to manufacturer’s protocol of In Situ Cell Death Detction Kit, TMR red (Roche, Mannhem, Germany).

Parallel experiments were repeated three times. Data was analyzed using SPSS 22.0 statistical software. And results are expressed as: mean ± standard deviation (SD). Mean was compared between groups using one Way-ANOVA. When the variance was homogeneous, LSD was used for multiple comparisons. Dunnett T3 method was applied for the unequal variances. Mean of two groups was compared with independent samples t test. A value of p < 0.05 indicated statistically significant difference.

An orthotopically implanted CRC model was used to characterize the anti-tumoral effects of sanguinarine in vivo [24]. The tumor weight was significantly reduced by both sanguinanine and Cisplatin in situ significantly (Fig. 1b and c). However, the body weight loss caused by sanguinarine was lower than that of caused by cisplatin (Fig. 1a). Increased TUNEL positive cells indicate that the cell death induced by sanguinarine can be defined as apoptosis (Fig. 1d and e). Western blot analysis showed that cleaved PARP and caspase 3 were increased significantly in sanguinarine-treated tissues (Fig. 1f). Further evidence indicated that cell proliferation was dampened by sanguinarine treatment (Fig. 1g and h). These results suggested that the growth of orthotopical implanted CRC was suppressed by sanguinarine via the induction of apoptosis.

Sanguinarine decreased the cell viability in dose- and time- dependent manner both in SW480 and HCT116 colorectal adenocarcinoma cells as monitored by 3-[4,5-dimethylthiazolyl-2]-2,5-diphenyl tetrazolium bromide staining at 24 h and 48 h (Fig. 2a). The number of Hoechst positive cells increased simultaneously upon treatment of sanguinanine (Fig. 2b). Chromatin condensation, rounding-up of the cells, cell shrinkage and extensive detachment of the cells from the cell culture substratum were observed by Hoechst 33,342 staining (Additional file 1: Figure S1 and Additional file 2: Figure S2). Flow cytometry showed that annexin V positive cells were significantly increased by sanguinarine treatment at 24 h and 48 h in a dose-dependent manner (Fig. 2c). Analysis of flow cytometry were presented in Additional file 3: Figure S3. To further define the cell death subroutines, cells were preincubated with a pan-caspase inhibitor, N-benzyloxycarbonyl-Val-Ala- Asp(OMe)-fluoromethyl ketone (Z-VAD-FMK). The annexin V positive cells were significantly reduced and the viability were elevated correspondingly by Z-VAD-FMK pretreatment (Fig. 2d and Additional file 4: Figure S4). These results suggested that sanguinarine induced apoptosis is the main cause for the reduced cell viability.

Apoptosis is a highly coordinated process that orchestrates a series of biochemical events. It recruits the pro-apoptotic members of the Bcl-2 family, such as Bax or Bak to control the formation of the mitochondrial apoptosis-induced channel (MAC) [26]. Sanguinarine induced the activation of Bax and downregulated the expression of Bcl-2 (Fig. 3a). The activation of Bax was accompanied by the release of cytochrome c from mitochondria to cytoplasm (Fig. 3b). The translocation of cytochrome c suggested the opening of MAC and mitochondrial outer membrane permeabilization (MOMP). DCFH-DA staining further showed that the ROS was significantly elevated by sanguirnarine, which can be effectively suppressed by L-N-acetyl-cysteine treatment (Fig. 3c and d).

JC-1 staining showed that the mitochondrial membrane potential (MMP) was significantly reduced by sanguinarine in SW480 cells (Fig. 3e and Additional file 5: Figure S5). In the cytosol, cytochrome c binds to Apaf-1, thereby triggering hierarchical activation of caspase-9 and caspase-3 [27]. The cleavage of caspase-9 and caspase-3, and subsequent cleavage of PARP were observed in SW480 cells treated with sanguinarine (Fig. 3f). The cell viability was restored by NAC pretreatment and the cleavage of caspase-3 and PARP was downregulated accordingly in cells treated with sanguinarine (Fig. 3g and h). Taken together, these results suggested that sanguinarine induces intrinsic apoptosis by triggering MOMP [24, 28].

To determine if the sanguinarine induced apoptosis is Bax-dependent, HCT116 Bax ^+/+ (wild type) and HCT116 Bax ^−/− cells were adopted [24]. Our results showed that the reduced cell viability by sanguinarine was significantly antagonized in HCT116 Bax ^−/− cells (Fig. 4a). Correspondingly, sanguinarine-induced apoptosis was significantly lowered in HCT116 Bax ^−/− cells (Fig. 4b). The apoptosis induced by sanguinarine in HCT116 Bax ^−/− cells was partly rescued by Bax transfection. But it wasn’t rescued by transfection of Bak (Additional file 6: Figure S6). The activation of Bax and downregulation of Bcl-2 were both abolished in HCT116 Bax ^−/− cells (Fig. 4c). JC-1 staining further confirmed that for HCT116 (WT), the mitochondrial membrane potential decreased more significantly than that of HCT116 Bax ^−/− (Additional file 7: Figure S7). Accordingly, the release of cytochrome c from mitochondria (Fig. 4d) and the cleavage of caspase-9, and casapase-3 were near totally abrogated in HCT116 Bax ^−/− cells (Fig. 4e), either. These results provided consolidate evidence that Bax is a key regulator in sanguinarine-induced apoptosis [29].

STRAP is up-regulated in 60% colon carcinomas and demonstrates oncogenic function in multiple tumors [11]. qRT-PCR (Fig. 5a) and western blot (Fig. 5b) were used to observe the effects of sanguinarine on the expression of STRAP. These results showed that sanguinarine significantly reduced its expression.

It has been previously reported that STRAP was a medium of bonding between maternal embryonic leucine zipper kinase (MELK) and TGF-β to regulate cell proliferation and apoptosis [9]. Interestingly, sanguinarine also decreased the expression of MELK in a time-dependent manner (Fig. 5b). Furthermore, the phosphorylation of STRAP and MELK was downregulated in a dose dependent manner by sanguinarine treatment (Fig. 5b). In vitro kinase assay showed that the kinase activity of MELK was significantly downregulated by sanguinarine in a dose-dependent manner (Fig. 5c). Our experiments showed that interaction between STRAP and MELK was significantly disrupted in SW480 cells upon sanguinarine treatment (Fig. 5d). These data suggested that sanguinarine lowers the expression and phosphorylation of STRAP and MELK and disassociates the interaction between them.

In breast cancer cells, MELK suppressed long isoform of Bcl-G (Bcl-G_L)-induced apoptosis [30]. To verify the role of MELK in Bax-mediated apoptosis, its expressions in HCT116 and HCT116 Bax ^−/− cells treated with sanguinarine were documented. Contrary to the significantly reduced expression in sanguinarine treated HCT116, the expression of MELK and STRAP in HCT116 Bax ^−/− cells was slightly elevated (Fig. 6a and Additional file 8: Figure S8). Western blot provide further evidence that sanguinarine decreased the phosphorylation and expression of STRAP and MELK in HCT116 but increased their expression slightly in HCT116 Bax ^−/− cells at the same time (Fig. 6b). To test if the specific association between STRAP and MELK was regulated by Bax, we probed the immunocomplex using anti-STRAP and anti-MELK antibody to immunoprecipitate each other in HCT116 and HCT116 Bax ^−/− cells, respectively. As expected, the interaction between STRAP and MELK was weakened by sanguinarine in HCT116 but was enhanced in HCT116 Bax ^−/− cells (Fig. 6c and d). Immunofluorescence confirmed the disassociation between STRAP and MELK in Bax positive cell treated with sanguinarine while showed strengthened association in HCT116 Bax ^−/− cells (Fig. 6e). These results suggested that the disrupted association between STRAP and MELK by sanguinarine was Bax-dependent.

The expression of Bax was elevated and the expressions of STRAP and MELK were decreased by sanguinarine treatment (Fig. 7a and b). When baited with MELK, immunoprecipitated STRAP were significantly reduced both in sanguinarine and cisplatin treated tumor tissues extracts (Fig. 7c). When probed with STRAP, the attenuated interaction between STRAP and MELK was also observed. The interaction between MELK and STRAP was also suppressed by cisplatin (Fig. 7c). Immunofluorescence test showed that the association between MELK and STRAP was weakened significantly by sanguinarine (Fig. 7d). Collectively, our in vivo experiments again confirmed that sanguinarine decreases the growth of CRC through downregulation and disassociation of STRAP and MELK.

To further explore the role of STRAP and MELK in tumorigenesis of CRC, the phosphorylation and expression of them were checked in CRC and paired normal tumor-adjacent colorectal tissues. No significant correlation was found between the expression level of MELK, STRAP and patients’ age, gender, and tumor size. The expression and phosphorylation of MELK and STRAP were significantly elevated in CRC specimens as compared to the paired normal tissues (Fig. 8a and b). Both MELK and STRAP were significantly correlated with tumor stages. Furthermore, the expression of MELK, but not STRAP, was associated with lymph node metastasis (Table 1). These results suggested that both STRAP and MELK have oncogenic potential and might be promising molecular target for development of therapy for CRC patients.