Cinobufagin induces acute promyelocytic leukaemia cell apoptosis and PML-RARA degradation in a caspase-dependent manner by inhibiting the β-catenin signalling pathway

Abstract Context Acute promyelocytic leukaemia (APL) is a malignant hematological tumour characterized by the presence of promyelocytic leukaemia–retinoic acid receptor A (PML-RARA) fusion protein. Cinobufagin (CBG) is one of the main effective components of toad venom with antitumor properties. However, only a few reports regarding the CBG treatment of APL are available. Objective We explored the effect and mechanism of action of CBG on NB4 and NB4-R1 cells. Materials and methods We evaluated the viability of NB4 and NB4-R1 cells treated with 0, 20, 40, and 60 nM CBG for 12, 24, and 48 h. After treatment with CBG for 24 h, Bcl-2 associated X (Bax), B-cell lymphoma 2 (Bcl-2), β-catenin, cyclin D1, and c-myc expression was detected using western blotting and real-time polymerase chain reaction. Caspase-3 and PML-RARA expression levels were detected using western blotting. Results CBG inhibited the viability of NB4 and NB4-R1 cells. The IC50 values of NB4 and NB4-R1 cells treated with CBG for 24 h were 45.2 nM and 37.9 nM, respectively. CBG induced NB4 and NB4-R1 cell apoptosis and PML-RARA degradation in a caspase-dependent manner and inhibited the β-catenin signalling pathway. Discussion and conclusion CBG induced NB4 and NB4-R1 cell apoptosis and PML-RARA degradation in a caspase-dependent manner by inhibiting the β-catenin signalling pathway. This study proposes a novel treatment strategy for patients with APL, particularly those with ATRA-resistant APL.


Introduction
Acute promyelocytic leukaemia (APL), also known as acute myeloid leukaemia type M3, is a haematopoietic malignancy characterized by the accumulation of large amounts of immature blood cells in haematopoietic tissues such as bone marrow. APL pathogenesis involves the translocation of chromosomes 15 and 17, which generates the promyelocytic leukaemia-retinoic acid receptor A (PML-RARA) fusion protein (Yilmaz et al. 2021). PML-RARA has dual functions of interfering with transcription regulation and destroying promyelocytic leukaemia homeostasis (Noguera et al. 2019). It can prevent normal differentiation of promyelocytic cells, block cell apoptosis, and stop at the promyelocytic stage, leading to APL occurrence. Moreover, PML-RARA has also become a significant diagnostic marker for APL (Wang et al. 2016). The degradation of PML-RARA fusion protein, which occurs via the autophagy-lysosomal (Isakson et al. 2010;Zeng et al. 2014), ubiquitin-proteasome (Rabellino et al. 2012), and caspase-dependent pathways (Wang et al. 2016), induces the differentiation of APL cells, thereby alleviating the disease (Kogan and Bishop 1999). Early, untreated APL is a dangerous, rapidly progressing, and lethal condition. Arsenic trioxide (ATO) and all-trans-retinoic acid (ATRA) significantly improve the prognosis of patients with APL. However, ATO and ATRA may cause adverse reactions, such as leukocytosis, infection, and heart, liver, and renal toxicity, and some patients develop ATRA resistance (Yamakawa et al. 2021). ATRA resistance affects the prognosis of patients, which is a serious obstacle in current clinical treatment. Therefore, highly efficient natural drugs with low toxicity that can overcome ATRA resistance are urgently needed to treat patients with APL.
In recent years, the treatment of cancer with natural products has gradually attracted attention . Cinobufagin (CBG) is extracted from toad venom, a traditional Chinese medicine with numerous functions, such as anti-inflammatory and antiviral functions, and is used as a cardiac glycoside. Recent studies have found that CBG plays an antitumor role in melanoma (Pan et al. 2019;Kim et al. 2020;, colorectal ) and nasopharyngeal (Pan et al. 2020) cancers, osteosarcoma (Ma et al. 2016), and other malignant tumours by inducing apoptosis (Pan et al. 2020;, blocking tumour angiogenesis , and arresting cell cycle (Pan et al. 2019(Pan et al. , 2020, and autophagy (Ma et al. 2016). However, only a few studies have reported the effects of CBG treatment on APL.
To determine whether CBG had an anti-APL effect on ATRA-sensitive and -resistant APL cell lines, we employed the NB4 (ATRA-sensitive) and NB4-R1 (ATRA-resistant) APL cell lines. NB4-R1 is a common ATRA-resistant APL cell model which has been used in various studies (Ge et al. 2014;Liang et al. 2019). In this study, we investigated the effects of CBG on ATRA-sensitive and -resistant APL cells with the aim of providing an experimental basis for the clinical treatment of ATRAresistant APL.

Cell culture
The ATRA-sensitive APL cell line (NB4 cells) was purchased from BeNa Culture Collection, and the ATRA-resistant APL cell line (NB4-R1 cells) was kindly provided by the Institute of Haematology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. The NB4 and NB4-R1 cells were grown in Roswell Park Memorial Institute 1640 medium (Gibco, Grand Island, NY, USA), supplemented with 10% foetal bovine serum (Gibco), and grown in an incubator with 37 C and 5% carbon dioxide.

Cell viability assay
Cells (5 Â 10 4 cells/mL, 200 lL) were seeded into 96-well plates and treated with different CBG doses for 24 h. Thereafter, 20 lL cell counting kit-8 (ApexBio, Houston, TX, USA) was added and incubated at 37 C for 1-4 h. The absorbance of the 96-well plate was measured at a wavelength of 450 nm using a microplate reader (Tecan, M€ annedorf, Switzerland).
Lactate dehydrogenase (LDH) activity assay NB4 and NB4-R1 cells were seeded at 1 Â 10 5 cells/mL into 6well plates, treated with different CBG concentrations, and incubated with serum-free media for 24 h. Thereafter, the cells were collected and centrifuged (3000 rpm, 5 min) to extract the supernatant. Supernatant LDH activity was measured using the LDH assay kit (Beyotime Biotechnology, Shanghai, China).

Hoechst 33258 staining
Hoechst 33258 staining kit (Beyotime) was used for apoptosis analysis. After 24 h of CBG treatment, cells were collected and stained with Hoechst 33258 staining solution for 5 min and placed on slides to observe under a fluorescence microscope (Nikon, Tokyo, Japan).

Flow cytometry analysis
To detect apoptosis using Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) double staining (Vazyme, Nanjing, Jiangsu, China), cells were inoculated at 10 6 cells/mL in Petri dishes containing different drug concentrations, incubated for 24 h, and collected. Annexin V-FITC and PI staining solution were added to each sample, which was incubated in the dark. Apoptosis was detected using flow cytometry (Beckman Coulter LTD, Brea, CA, USA).

Western blotting
Cells were inoculated at 10 6 cells/mL in Petri dishes containing different drug concentrations, incubated for 24 h, and then collected and lysed with lysis buffer (radioimmunoprecipitation assay buffer: phenylmethylsulfonyl fluoride: protein phosphatase inhibitor ¼ 100:1:1). The protein concentration was determined using a bicinchoninic acid (BCA) protein concentration determination kit (Beyotime). Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane (Micon, Miller, MA, USA), which was blocked with Tris-buffered saline with Tween (TBST) containing 5% bovine serum albumin (BSA) for 1 h, washed with TBST, and incubated with primary antibody at 4 C overnight. Thereafter, the membrane was incubated with secondary antibody in TBST for 1 h and washed, and the protein bands were visualized using a gel imager (Bio-Rad, Hercules, CA, USA).

JC-1 staining
Mitochondrial membrane potential (MMP) was measured using an MMP measurement kit (JC-1, Beyotime). The cells were collected after centrifugation, washed, and incubated with JC-1 staining solution at 37 C for 30 min, followed by fluorescence microscopy (Nikon).

Reactive oxygen species (ROS) activity detection
The cells were collected and incubated with 2 0 , 7 0 -dichlorodihydrofluorescein diacetate (Beyotime) for 30 min, and ROS activity was detected using a fluorescence microscope (Nikon) or fluorescence microplate reader (Tecan) with excitation at 488 nm and emission at 525 nm.

20s proteasome activity
After CBG treatment for 24 h, cells were collected, washed, and lysed with lysis buffer without protease inhibitors. Protein concentration was measured using a BCA kit (Beyotime). The same amounts of protein and substrate (Suc-Leu-Leu-Val-Tyr-AMC) were co-incubated at 37 C for 1 h, and AMC release was detected using a fluorescence microplate reader (Tecan) with excitation at 355 nm and emission at 460 nm.

Ca 21 activity assay
After 24 h of CBG treatment, cells were collected, washed, and stained with 4 lM Fluo-4 (Beyotime) staining solution at 37 C for 30 min in the dark. Thereafter, the cells were washed with PBS and incubated for 20 min. Ca 2þ activity was then detected using a fluorescence microscope (Nikon) or fluorescence microplate reader (Tecan) with excitation at 488 nm and emission at 515 nm.

Real-time polymerase chain reaction (RT-PCR)
In an RNA-free environment, RNA was extracted from cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The RNA was then reverse transcribed using RT SuperMix for the qPCR reverse transcription kit (Vazyme, Nanjing, Jiangsu, China). Thereafter, cDNA was amplified using the ChamQ Universal SYBR qPCR Master Mix reagent. The gene expression of each sample was normalized to glyceraldehyde 3-phosphate dehydrogenase (GADPH) expression as an internal control, and the expression of each target gene was calculated using the 2 ÀDDCT method. The primers used for amplification are as follows: The data are expressed as mean ± SD (n ¼ 6). Compared with solvent group, Ã P < 0.05, ÃÃ P < 0.01, ÃÃÃ P < 0.001. ns: no significant difference.

Statistical analysis
Statistical analysis was performed with GraphPad Prism 9 software (GraphPad Software Inc. CA, USA). All data are expressed as mean ± standard deviation. t-Test was used to compare the two groups, and one-way analysis of variance (ANOVA) was used to compare multiple groups. P < 0.05 was considered to indicate significance.

CBG inhibited the viability of NB4 and NB4-R1 cells
The structural formula of CBG is shown in Figure 1(A). The viability of NB4 and NB4-R1 cells was first detected after 12, 24, and 48 h of CBG treatment. Notably, CBG inhibited the viability of NB4 and NB4-R1 cells in a time-and dose-dependent manner (Figure 1(B,C)). CBG (20, 40, and 60 nM) was co-incubated with NB4 and NB4-R1 cells for 24 h, resulting in an increase in LDH release (Figure 1(D,E)). Furthermore, CBG induces myocardial toxicity (Xiong et al. 2019). However, we found that CBG had no significant effect on the myocardial cell line (H9C2 cells) after 24 h of treatment at CBG concentrations of 0-60 nM (Figure 1(F)).
CBG induced apoptosis and inhibited PML-RARA expression in NB4 and NB4-R1 cells Cleaved caspase-3 was detected after the administration of CBG. (Figure 2(A,B)). Bax and Bcl-2 are key apoptotic proteins of the Bcl family, and the Bax/Bcl-2 ratio is an important indicator of apoptosis. Western blotting results revealed that the Bax/Bcl-2 ratio was significantly increased after CBG treatment ( Figure  2(A-D)). Following CBG administration, NB4 and NB4-R1 cells exhibited nucleus fragmentation (Figure 2(E,F)) and increased apoptosis (Figure 2(G,H)). Z-VAD-FMK, a broad-spectrum caspase inhibitor, inhibited cleaved caspase-3 (Figure 3(A,B)) and increased cell viability after CBG treatment (Figure 3(C,D)). We also found that CBG treatment inhibited PML-RARA expression (Figure 2(A,B)). These data suggested that CBG promoted apoptosis in a caspase-dependent manner and inhibited PML-RARA expression in NB4 and NB4-R1 cells.

CBG-induced apoptosis in NB4 and NB4-R1 cells was accompanied by mitochondrial damage
Apoptosis is often accompanied by mitochondrial damage, and the decrease in MMP is a sign of mitochondrial damage (Zhu et al. 2021). To further explore whether CBG caused mitochondrial damage, we measured MMP, interestingly, MMP decreased ( Figure 4(A,B)). Mitochondria are the main sites of ROS production, and increased ROS might reflect mitochondrial damage. Further experiments revealed a large amount of ROS in NB4 and NB4-R1 cells after CBG administration (Figure 4(C-F)). These results suggested that CBG induced mitochondrial damage in NB4 and NB4-R1 cells.

CBG induced endoplasmic reticulum (ER) stress in NB4 and NB4-R1 cells
ER stress is often associated with apoptosis (Fern andez et al. 2015). ER stress modulates three pathways: inositol-requiring enzyme 1 (IRE 1), PERK, and ATF6 (Oh et al. 2021). Western blotting revealed that CBG activated the PERK-eIF2a-ATF4 pathway ( Figure 5(A,B)). Moreover, the expression levels of CHOP, an ER stress marker, were significantly increased ( Figure  5(C-F)). Maintaining calcium homeostasis is one of the main functions of the ER, and ER stress often affects intracellular Ca 2þ homeostasis (Kim and Kim 2018). Notably, we detected intracellular Ca 2þ overload using a fluorescence microplate reader ( Figure 6(A,B)) and fluorescence microscope ( Figure  6(C,D)). These results suggested that CBG induced ER stress in NB4 and NB4-R1 cells.

CBG-induced PML-RARA degradation was caspase-dependent
APL pathogenesis involves a chromosomal translocation, which generates the PML-RARA fusion protein (Yilmaz et al. 2021). Based on the above-mentioned experimental results, CBG effectively inhibited PML-RARA expression (Figure 2(A,B)). To explore the mechanism by which CBG inhibited the expression of PML-RARA fusion protein, we used CHX, a broad-spectrum protein synthesis inhibitor, in combination with CBG. Intriguingly, the combination of CHX and CBG inhibited PML-RARA expression significantly more strongly than CHX alone did (Figure 7(A,B)). PML-RARA is degraded via three pathways: the ubiquitin-proteasome (Rabellino et al. 2012), autophagy-lysosome (Isakson et al. 2010), and caspase-dependent (Wang et al. 2016) pathways. To further explore the mechanism by which CBG promoted PML-RARA fusion protein degradation, we conducted proteasome activity experiments (Figure 7(C,D)) and western blotting for autophagy-related proteins LC3 and p62 (Figure 7(E,F)). Notably, we found that LC3 II/I expression levels did not increase in NB4 and NB4-R1 cells after CBG treatment (Figure 7(E,F)). Additionally, p62 degradation may be related to caspase . Therefore, we added Z-VAD-FMK and confirmed that p62 degradation was related to caspase (Figure 7(G,H)), which excluded autophagy, indicating that CBG promoted PML-RARA degradation independently of the ubiquitin-proteasome and lysosome pathways. Subsequently, we found that the CBG-induced degradation of PML-RARA fusion protein was significantly inhibited after the addition of Z-VAD-FMK (Figure 7(G,H)). These data demonstrated that CBG    The cells were incubated with CHX (20 lM) for 2 h and then treated with 60 nM CBG for 24 h to detect the PML-RARA by western blotting. The results are expressed as mean ± SD (n ¼ 3). Compared with the CBG group with a specified dose, Ã P < 0.05, ÃÃ P < 0.01, ÃÃÃ P < 0.001. (C, D) Cells were incubated with CBG (0-60 nM) for 24 h, and then collected for 20S protease activity determination. (E, F) The expression levels of LC3 and p62 were detected by western blotting. Compared with solvent group, Ã P < 0.05, ÃÃ P < 0.01, ÃÃÃ P < 0.001, ns: no significant difference. (G, H) Western blotting was used to detect the expression of p62 and PML-RARA after adding Z-VAD-FMK. Results are expressed as mean ± SD (n ¼ 3). Compared with the CBG group with a specified dose, Ã P < 0.05, ÃÃ P < 0.01, ÃÃÃ P < 0.001.
induced PML-RARA fusion protein in a caspase-dependent manner.

CBG inhibited the b-catenin pathway
The b-catenin signalling pathway is extensively involved in various intracellular physiological functions such as cell invasion, migration , differentiation ) and apoptosis . Our data showed that CBG inhibited the expression of b-catenin and its downstream target protein c-myc and cyclin D1 (Figure 8(A-D)). Immunofluorescence assay also revealed that b-catenin expression levels significantly decreased after 24 h of treatment with 60 nM CBG ( Figure  8(E,F)). We found that CBG alone increased the Bax/Bcl-2 ratio and cleaved caspase-3 expression levels and inhibited PML-RARA expression (Figure 2(A,B)). However, CBG in combination with LiCl, a b-catenin activator, reversed the increase in Bax/Bcl-2 ratio and cleaved caspase-3 expression levels and the CBG-inhibited PML-RARA expression (Figure 8(G,H)). Therefore, the b-catenin pathway was likely involved in CBGinduced apoptosis and PML-RARA degradation in NB4 and NB4-R1 cells.

Discussion
APL is characterized by a chromosomal translocation which generates the PML-RARA fusion protein (Fasan et al. 2017). Early, untreated APL is a rapidly progressing condition with a high fatality rate. The combination of ATO and ATRA is often used as a clinical treatment for APL. However, adverse reactions, such as leukocytosis and infection, caused by ATO and ATRA may worsen the prognosis of patients with APL, and some patients are prone to relapse and ATRA resistance. Therefore, a highly efficient natural substance with minimal toxicity that can overcome ATRA resistance is required. CBG is an active substance in toad venom, a traditional Chinese medicine, that has been widely The mRNA levels of b-catenin, c-myc and cyclin D1 were detected by RT-PCR. Compared with solvent group, Ã P < 0.05, ÃÃ P < 0.01, ÃÃÃ P < 0.001. (E, F) CBG treated NB4 and NB4-R1 cells for 24 h, the expression of b-catenin protein was determined by immunofluorescence assay. (G, H) After preincubation with LiCl, a b-catenin activator, for 2 h, CBG was treated with 24 h to detect the protein expression by western blotting. The results are exhibited as mean ± SD (n ¼ 3). Compared with the CBG group with a specified dose, Ã P < 0.05, ÃÃ P < 0.01, ÃÃÃ P < 0.001. used to treat various cancers. However, only a few reports on CBG treatment of APL are available. This study demonstrated the in vitro effects and mechanisms of action of CBG on APL.
Previous studies have shown that CBG treats cancer by promoting apoptosis (Zhang L et al. 2020), autophagy (Ma et al. 2016) and cell cycle arrest (Pan et al. 2019). First, we found that CBG inhibited the viability of NB4 and NB4-R1 cells in a timeand dose-dependent manner. CBG is cardiotoxic (Xiong et al. 2019), therefore, we performed experiments with a myocardial cell line (H9C2 cells) and found no significant effects at CBG concentrations of 0-60 nM. Our data suggested that CBG inhibited the viability of NB4 and NB4-R1cells without cardiotoxic at these doses.
Mitochondrial damage is an important feature of apoptosis. Moreover, decreased MMP is a common sign of mitochondrial damage (Zhu et al. 2021). Mitochondria are the main sites of ROS, and massive ROS production may reflect mitochondrial damage. In the present study, CBG induced caspase-dependent apoptosis in NB4 and NB4-R1 cells. We experimentally confirmed that CBG induced ROS accumulation and mitochondrial damage.
Several natural products that induce ER stress and apoptosis, thereby exerting antitumor effects, have been discovered in recent years (Kim and Kim 2018). ER stress caused by intracellular misfolded and unfolded proteins can activate unfolded protein response via three pathways: the IRE 1, PERK, ATF6 (Pak et al. 2019). Simultaneously, ER stress is accompanied by an imbalance in cytoplasmic calcium homeostasis. Notably, we found that CBG activated the PERK-eIF2a-ATF4 pathway, causing an increase in intracellular Ca 2þ , which confirmed that CBG triggered ER stress in NB4 and NB4-R1 cells. ER stress-induced apoptosis is mainly mediated by CHOP which is activated by the PERK-eIF2a-ATF4 axis. In the present study, CHOP expression was increased. Particularly, CHOP regulates Bax and Bcl-2 expression (Alshammari et al. 2021;Chen et al. 2021;Hardy et al. 2021;Zheng et al. 2021), activates downstream caspase, and induces apoptosis (Hsu et al. 2018;Huang et al. 2018;Isobe et al. 2019). In the present study, we found that the Bax/Bcl-2 ratio increased following CBG administration.
APL is characterized by the presence of PML-RARA fusion protein (Yilmaz et al. 2021). In the present study, PML-RARA was inhibited. We used CHX, a broad-spectrum protein synthesis inhibitor, to inhibit PML-RARA synthesis and confirmed that the combination of CBG and CHX significantly enhanced the inhibition of PML-RARA expression compared to CHX alone, suggesting that CBG might promote PML-RARA degradation. Particularly, PML-RARA is degraded via the ubiquitin-proteasome (Zhu et al. 1999;Kopf et al. 2000;Nasr et al. 2009), autophagy-lysosomal pathway and caspasedependent (Zhu et al. 1999;Nasr et al. 2009) pathways. Noteworthy, we found that CBG induced apoptosis and that caspase was a key target of CBG; therefore, we speculated that CBG-induced PML-RARA degradation might be related to caspase activation. To explore the pathway by which CBG induced the degradation of PML-RARA fusion protein, we performed a proteasome activity assay and found that CBG effectively inhibited proteasome activity. Additionally, we performed western blotting and detected the expression of autophagy-related proteins, LC3, and p62. Generally, LC3 II/I expression increased during autophagy, reflecting autophagosome formation (Kabeya et al. 2000), and the expression of p62, an autophagy degradation substrate (Islam et al. 2018), was decreased. We found that LC3 II/I expression was not increased, indicating the absence of autophagy. Interestingly, p62 expression was decreased; in particular, the cleavage of p62 is related to caspase activation and is involved in apoptosis . To further explore PML-RARA degradation, we treated the cells with Z-VAD-FMK, a broad-spectrum caspase inhibitor, which inhibited PML-RARA degradation. These results suggested that CBG degraded the PML-RARA fusion protein in a caspase-dependent manner, but not via autophagy-lysosome and ubiquitin-proteasome pathways. Interestingly, the decrease in p62 expression was reversed after the addition of Z-VAD-FMK; therefore, we speculated that p62 degradation was autophagy-independent and possibly caspase-dependent.
The b-catenin signalling pathway is involved in various physiological activities, such as differentiation, invasion, migration, and apoptosis ). According to our results, CBG inhibited the expression of b-catenin and its downstream factors, c-myc and cyclin D1. Notably, LiCl, a b-catenin activator, reversed the CBG-induced increase in the Bax/Bcl-2 ratio and in cleaved caspase-3, as well as the CBG-induced decrease in PML-RARA expression. Nevertheless, determining whether b-catenin induced apoptosis by regulating ER stress or mitochondrial damage requires further investigation.

Conclusion
This study indicated that CBG induced NB4 and NB4-R1 cell apoptosis and PML-RARA degradation in a caspase-dependent manner, and that the b-catenin signalling pathway played a crucial role in this process (Figure 9). Therefore, CBG might be a potential candidate treatment for APL and ATRA-resistant APL.