HO-3867 Induces Apoptosis via the JNK Signaling Pathway in Human Osteosarcoma Cells

Metastatic osteosarcoma often results in poor prognosis despite the application of surgical en bloc excision along with chemotherapy. HO-3867 is a curcumin analog that induces cell apoptosis in several cancers, but the apoptotic effect and its mechanisms on osteosarcoma cells are still unknown. After observing the decrease in cellular viability of three human osteosarcoma U2OS, HOS, and MG-63 cell lines, and the induction of cellular apoptosis and arrest in sub-G1 phase in U2OS and HOS cells by HO-3867, the human apoptosis array showed that heme oxygenase (HO)-1 and cleaved caspase-3 expressions had significant increases after HO-3867 treatment in U2OS cells and vice versa for cellular inhibitors of apoptosis (cIAP)1 and X-chromosome-linked IAP (XIAP). Western blot analysis verified the results and showed that HO-3867 activated the initiators of both extrinsic caspase 8 and intrinsic caspase 9, and significantly increased cleaved PARP expression in U2OS and HOS cells. Moreover, with the addition of HO-3867, ERK1/2, and JNK1/2 phosphorylation were increased in U2OS and HOS cells. Using the inhibitor of JNK (JNK in 8), HO-3867’s increases in cleaved caspases 3, 8, and 9 could be expectedly suppressed, indicating that JNK signaling is responsible for both apoptotic pathways, including extrinsic and intrinsic, in U2OS and HOS cells caused by HO-3867. Through JNK signaling, HO-3867 has proven to be effective in causing both extrinsic and intrinsic apoptotic pathways of human osteosarcoma cells.


Introduction
Osteosarcoma, the most common tumor of primary malignant bone tumor, is mostly found in children and adolescents with a peak of incidence at 11-15 years and about six in every million children [1,2]. Typically, complete surgical en bloc excision, or extensive amputation of the affected area, was the most common type of treatment when attempting a complete radical excision, but it did not provide a good prognosis. Fortunately, chemotherapy has become a vital part for the treatment of osteosarcoma [3,4]. Using a combination of chemotherapy and surgery, long-term survival rates of diagnosed patients have improved to approximately 68-75% at 5 years for diagnosed patients with localized tumor [2,5]. Unfortunately, due to its metastatic ability, lung transfer osteosarcoma is still responsible for undesirable outcomes and fatalities [6,7]. To prevent this, several approaches to the development of new compounds containing anticancer mechanisms such as cytotoxic and antimetastatic activities need to be developed.

Cell Culture and HO-3867 Treatment
All the human osteosarcoma U2OS, HOS, and MG-63 cells were purchased from the FIRDI (Hsinchu, Taiwan). The U2OS cells were cultured in DMEM and supplemented with 10% FBS, 5 mL of glutamine, and 1% penicillin. The HOS and MG-63 cells were cultured in DMEM and supplemented with 10% FBS, 1% penicillin/streptomycin, and 5 mL glutamine. The cell cultures were maintained at 37 • C in a humidified atmosphere of a 5% CO 2 incubator. HO-3867 was obtained from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan).

Microculture Tetrazolium Colorimetric (MTT) Assay
To gather data regarding the effects of apoptosis caused by HO-3867 on osteosarcoma cells, we extracted cells from 8.5 × 10 4 /well of U2OS, of HOS, and of MG-63 and applied different HO-3867 concentrations (0, 2,4,8,16, and 32 µM) of for 24 h within 24-well plates. After completing the exposure period, the media was separated and the cells were washed using phosphate-buffered saline. Then, new medium was added and the cells were then incubated using MTT [26,27]. Following solubilization with isopropanol, the viable cell number, directly proportional to the production of formazan, was measured spectrophotometrically at 563 nm.

Flow Cytometric Analysis
We can determine the phases of the cell cycle affected by HO-3867, as well as several other cellular components such as DNA, using flow cytometry analysis on U2OS and HOS cells. Summarily, we plated 8 × 105 U2OS and 6 × 10 5 HOS cells in 6 cm dishes and placed them in experimental concentration range (0, 2, 4, 8, and 16 µM) of HO-3867 for 24 h. After propidium iodide (PI) staining, 2 × 105 U2OS and HOS cells were placed in an Eppendorf tube to analyze the cell cycle using a BD AccuriTM C6 Plus personal flow cytometer [28,29].

Annexin V-FITC Apoptosis Staining Assay
After going through apoptosis, cells would translocate membrane phospholipid phosphatidylserine molecules from the inner to the surface layer of the plasma membrane. Using Annexin V, a conjugated fluorescent protein with a high affinity for the translocated molecules, we stained the phospholipid phosphatidylserine molecules, which were now exposed externally, making it easier to identify apoptosis in earlier stages than other assays, such as PI staining, which were based on nuclear changes. We treated approximately 8 × 105 U2OS and 6 × 105 HOS cells with experimental concentration range of HO-3867 for 24 h in one 6 cm plate. Following that, trypsinization was used to harvest viable cells along with floating nonviable cells. Following the protocol given by the manufacturer (BD Biosciences, San Jose, CA, USA), FITC Annexin V Apoptosis Detection Kit I was administered, followed by the analysis of the cell cycle through flow cytometry. Annexin V-FITC apoptosis staining was used in conjugation with PI staining to determine apoptosis from necrosis [29,30].

Human Apoptosis Array
To understand the effects of induced apoptosis, we followed the manufacturer's protocols and used a Human Apoptosis Array Kit to define protein lysates from a vehicle-8 µM HO-3867-containing 2.4 × 106 U2OS cells that were treated for 24 h. In total, 35 proteins related to apoptosis were detected. The proteins were placed on a nitrocellulose membrane, detected with biotinylated detection antibodies, and then finally visualized through using chemiluminescent detection reagents.

Protein Extraction and Western Blot Analysis
We treated 8 × 105 U2OS and 6 × 105 HOS cells within a cell plate with experimental concentration range of HO-3867. The total cell lysates of U2OS and HOS cells were gathered and had their proteins extracted. We performed Western blot analysis using the primary antibodies against both uncleaved and cleaved forms of caspases 3, 8, and 9, as well as the antibodies for both unphosphorylated and phosphorylated forms of the MAPKs. For the antibody dilution, all antibodies were 1:1000 dilutions, except HO-1 antibody (1:5000 dilution). Horseradish peroxidase goat anti-rabbit and anti-mouse was then used for incubation before densitometry was used to measure the intensity [26,27]. After the intensity of each band was measured by densitometry, the relative intensities were calculated by normalizing to β-actin (1:1000 dilution; Santa Cruz Biotechnology, Inc., Dallas, TX, USA).

Statistical Analyses
The data from experiments went through statistical calculations performed by oneway analysis of variance (ANOVA) along with post hoc Tukey tests for more than two groups with equal sample sizes per group. Experiments were performed as independent and at least in triplicate experiments.

HO-3867 Induces Cell Death in Human Osteosarcoma U2OS, HOS, and MG-63 Cells
The chemical structures of curcumin and curcumin analog HO-3867 were drawn in Figure 1A. To define cytotoxicity of HO-3867 on osteosarcoma U2OS, HOS, and MG-63  Figure 1C,D, treatment of cells with HO-3867 for 24 h significantly decreased the proportion of viable cells in a concentration-dependent manner. Therefore, we picked the U2OS and HOS cell lines and used the experimental concentration range (0, 2, 4, 8, and 16 µM) for HO-3867 to explore its anticancer properties in the subsequent experiments.

HO-3867 Induces Cell Death in Human Osteosarcoma U2OS, HOS, and MG-63 Cells
The chemical structures of curcumin and curcumin analog HO-3867 were drawn in Figure 1A. To define cytotoxicity of HO-3867 on osteosarcoma U2OS, HOS, and MG-63 cells, an MTT assay was performed. After treatment with HO-3867 for 24 h, U2OS, HOS, and MG-63 cells' viability in concentrations of 2,4,8,16, and 32 μM of HO-3867 was significantly unlike that of controls (0 μM) and showed dose-dependently (U2OS: p < 0.001; HOS: p < 0.001; MG-63: p < 0.001). ( Figure 1B) After 24 h of HO-3867 (4, 8, and 16 μM) treatment, cytotoxicity in U2OS and HOS cells had dose-dependent increases, and their half maximal inhibitory concentrations (IC50) of HO-3867 were 6.91 μM in U2OS cells, 7.60 μM in HOS cells, and 12.24 μM in MG-63 cells. Moreover, cell proliferation was assessed by using the CCK-8 method in U2OS and HOS cells. As shown in Figure 1C,D, treatment of cells with HO-3867 for 24 h significantly decreased the proportion of viable cells in a concentration-dependent manner. Therefore, we picked the U2OS and HOS cell lines and used the experimental concentration range (0, 2, 4, 8, and 16 μM) for HO-3867 to explore its anticancer properties in the subsequent experiments.

HO-3867 Induces Cell Apoptosis and Arrest in the Sub-G1 Phase of U2OS and HOS Cells
To investigate the unknown mechanisms of HO-3867 inhibition of U2OS and HOS cell proliferation, flow cytometry was performed to examine the cell cycle. After being

HO-3867 Induces Cell Apoptosis and Arrest in the Sub-G1 Phase of U2OS and HOS Cells
To investigate the unknown mechanisms of HO-3867 inhibition of U2OS and HOS cell proliferation, flow cytometry was performed to examine the cell cycle. After being stained with PI, flow cytometry showed that 8 µM of HO-3867 drastically increased sub-G1 phase cell cycle accumulation from 1.7% to 48.7% in U2OS cells and 3.9% to 35.1% in HOS cells, suggesting that HO-3967 causes sub-G1 phase arrest in U2OS and HOS cells (Figure 2A-C).
Additionally, detecting apoptosis at earlier stages, before gross morphological changes, is crucial for understanding the signaling pathways of programmed cell death. To verify whether or not the suppressive effects of HO-3867 on cell growth were caused by apoptosis and not others such as necrosis, Annexin V-FITC/PI apoptosis assay was used. Using both Annexin V-FITCH with PI staining, flow cytometry ensured that HO-3867 induced apoptosis of U2OS and HOS cells ( Figure 3A,B). stained with PI, flow cytometry showed that 8 μM of HO-3867 drastically increased sub-G1 phase cell cycle accumulation from 1.7% to 48.7% in U2OS cells and 3.9% to 35.1% in HOS cells, suggesting that HO-3967 causes sub-G1 phase arrest in U2OS and HOS cells (Figure 2A-C). Additionally, detecting apoptosis at earlier stages, before gross morphological changes, is crucial for understanding the signaling pathways of programmed cell death. To verify whether or not the suppressive effects of HO-3867 on cell growth were caused by apoptosis and not others such as necrosis, Annexin V-FITC/PI apoptosis assay was used. Using both Annexin V-FITCH with PI staining, flow cytometry ensured that HO-3867 induced apoptosis of U2OS and HOS cells ( Figure 3A,B).

HO-3867 Increases the Cleaved Caspase 3 and Heme Oxygenase (HO)-1 Expression but Decreases XIAP and cIAP1 Expression in U2OS and HOS Cells
To demonstrate the mechanisms of apoptosis in U2OS cells caused by HO-3867, the human apoptosis array kit was used for determining apoptosis-related proteins. The human apoptosis array was performed on the U2OS cells that were treated with HO-3867 for 24 h, and the results showed increases in the cleaved caspase 3 and HO-1 proteins and decreases in cIAP1 and XIAP proteins ( Figure 4A). To confirm the findings, Western blotting and quantitative analysis showed a significant increase in HO-1 (U2OS: p < 0.001; HOS: p < 0.001) but significant decreases in XIAP and cIAP1 in U2OS (XIAP: p < 0.001; cIAP1: p < 0.001) and HOS cells (XIAP: p < 0.001; cIAP1: p < 0.001) ( Figure 4B).

HO-3867 Increases the Cleaved Caspase 3 and Heme Oxygenase (HO)-1 Expression but Decreases XIAP and cIAP1 Expression in U2OS and HOS Cells
To demonstrate the mechanisms of apoptosis in U2OS cells caused by HO-3867, the human apoptosis array kit was used for determining apoptosis-related proteins. The human apoptosis array was performed on the U2OS cells that were treated with HO-3867 for 24 h, and the results showed increases in the cleaved caspase 3 and HO-1 proteins and decreases in cIAP1 and XIAP proteins ( Figure 4A). To confirm the findings, Western blotting and quantitative analysis showed a significant increase in HO-1 (U2OS: p < 0.001; HOS: p < 0.001) but significant decreases in XIAP and cIAP1 in U2OS (XIAP: p < 0.001; cIAP1: p < 0.001) and HOS cells (XIAP: p < 0.001; cIAP1: p < 0.001) ( Figure 4B). Within the tested proteins in the human apoptosis array, cleaved caspase 3 increased the most, meaning that the caspase 3 effector is the one responsible for dismantling U2OS and HOS cells. To identify the underlying effects of the caspase cascade caused by HO-3867, Western blotting was used to discover the effector caspase 3 as well as its upstream initiator caspases 8 and 9, and their cleaved forms. U2OS and HOS cells were then treated with experimental concentration range of HO-3867 for 24 h, the results dose-dependently showed fewer levels of pro-caspases 3, 8, and 9 dose-dependently (U2OS: p < 0.001; p < 0.001; p < 0.001; HOS: p < 0.001; p < 0.001; p < 0.001) and more expressions of cleaved caspases 3, 8, and 9 within higher concentrations (U2OS: p < 0.001; p < 0.001; p < 0.001; HOS: p < 0.001; p < 0.001; p < 0.001) ( Figure 5A,B). We then discovered that HO-3867 activates extrinsic caspase 8 and intrinsic caspase 9 along with the downstream effector, caspase 3, to causes apoptosis in U2OS and HOS cells.
Within the tested proteins in the human apoptosis array, cleaved caspase 3 increased the most, meaning that the caspase 3 effector is the one responsible for dismantling U2OS and HOS cells. To identify the underlying effects of the caspase cascade caused by HO-3867, Western blotting was used to discover the effector caspase 3 as well as its upstream initiator caspases 8 and 9, and their cleaved forms. U2OS and HOS cells were then treated with experimental concentration range of HO-3867 for 24 h, the results dose-dependently showed fewer levels of pro-caspases 3, 8, and 9 dose-dependently (U2OS: p < 0.001; p < 0.001; p < 0.001; HOS: p < 0.001; p < 0.001; p < 0.001) and more expressions of cleaved caspases 3, 8, and 9 within higher concentrations (U2OS: p < 0.001; p < 0.001; p < 0.001; HOS: p < 0.001; p < 0.001; p < 0.001) ( Figure 5A,B). We then discovered that HO-3867 activates extrinsic caspase 8 and intrinsic caspase 9 along with the downstream effector, caspase 3, to causes apoptosis in U2OS and HOS cells.

HO-3867 Activates Apoptotic Processes via the JNK-Signaling Pathway in U2OS and HOS Cells
MAPK pathways play an important role in regulating apoptosis by chemotherapeutic drugs as well as also being the upstream signaling of caspases 3, 8, and 9. To investigate further molecular mechanisms, Western blot analysis was then administered. As displayed in Figure 6A-D, HO-3867 was shown in increasing the phosphorylation of ERK 1/2 as well as JNK 1/2 dose-dependently within U2OS (p < 0.001; p < 0.001) and HOS cells (p < 0.001; Pharmaceutics 2022, 14, 1257 9 of 14 p < 0.001), which indicates that HO-3867 activated the phosphorylation of the ERK 1/2 and JNK 1/2 pathways in the osteosarcoma cells. Yet, the phosphorylation of p38 showed inconsistent decreases after the application of HO-3867 in U2OS (p < 0.001) and HOS cells (p < 0.001). To identify whether the activated phosphorylation of ERK 1/2 and JNK 1/2 by HO-3867 would affect the intrinsic and extrinsic processes of U2OS and HOS cells through caspases 3, 8, and 9, a combination of the inhibitors of ERK 1/2 (U0126) and JNK 1/2 (JNK-IN-8) with or without treatment was used in Western blot analysis. Cleaved caspases 3, 8, and 9 were expectedly activated by 8 µM of HO-3867 (p < 0.001, p < 0.001, and p < 0.001) ( Figure 7A,B). Additionally, the inhibitors of JNK1/2 significantly repressed the increase in cleaved caspases 3, 8, and 9 caused by the treatment of HO-3867 in U2OS and HOS cells. However, the inhibitor of ERK did not show suppressive effects on the increased levels of caspases 3, 8, and 9 caused by the treatment of HO-3867. These findings suggest that the JNK1/2 pathway is critical in the HO-3867-mediated apoptosis of extrinsic and intrinsic pathways as well as the downstream effector in U2OS and HOS cells.

Discussion
The high mortality and the main cause of most treatment failures rates of osteosarcoma is its highly metastatic potential [4,31] and failed chemotherapeutic treatments can result in the development of resistance within the tumor for future attempts of treatments [15,16,32]. Hence, to minimize the possibility of failed treatments, we conducted experiments and explored the underlying mechanisms of HO-3867, which is a synthesized curcumin analog for improving bioavailability compared to its original counterpart of curcumin, known for its apoptotic mechanisms at the molecular level [32]. Several studies have mentioned that curcumin is already considered to have a good anticancer effect [33,34]. In this study, we studied and demonstrated that HO-3867 possesses anticancer properties in human osteosarcoma.
HO-3867 has previously been known to cause apoptosis to cancer cells through the targeting of several key growth-regulatory proteins such as the Janus kinase (JAK) as well as STAT3 pathway to cause apoptosis among oral, ovarian, endometrial, and pancreatic cancers [35][36][37][38]. The compound has also been researched in terms of its ability to repress migration and invasion activity [39]. Moreover, HO-3867 has been shown to regulate the expression of FAS, FAK, and VEGF in order to suppress metastasis of ovarian carcinoma cells [40]. Overall, HO-3867 is versatile and has demonstrated an ability to initiate apoptosis or regulate metastasis in cancer cells by regulating various proteins in cancers.
Knowing from previous studies that HO-3867 is associated with apoptosis through various pathways in other cancer types, we focused on the effects of HO-3867 of various concentrations on various cell lines of human osteosarcoma. Results from flow cytometry showed that HO-3867 initiated apoptosis and decreased cell viability in human osteosarcoma U2OS, HOS, and MG-63 cells through the sub-G1 phase. As apoptosis can be initiated through various ways, we narrowed down the pathway after discovering an increase in HO-1 and a decrease in cIAP1 and XIAP. This is a crucial step in allowing us to identify that MAPK was utilized by HO-3867 for the results previously mentioned. Moreover, HO-3867 is widely considered as a selective STAT3 inhibitor [18,24]. Numerous studies have reported that STAT3 inhibitor exerts anticancer activity via different MAPK pathways [41][42][43]. However, the detailed mechanisms of how HO-3867 activates ERK 1/2 and diminishes p38 in U2OS and HOS cells still require further investigation.
Although p53 is an important tumor suppressor gene, it is one of the most frequently mutated genes in cancer, implicated in more than half of all human cancers [44,45]. Mutant p53 (p53 MT ) loses the activity of wild-type p53 or expresses mutant proteins to inhibit the activity of the genome-guarding function through multiple mechanisms, which depend on different types of cancer and cell lines, even in the same cancer [44][45][46]. However, wild-type p53 (p53 WT ) is paradoxically retained in certain types or cell lines of cancers, such as p53 WT U2OS and p53 MT HOS cells [45,47]. HO-3867 covalently binds to mutant p53 to convert the mutant p53 protein to transcriptionally activate the wild-type p53 anticancer genetic response [44], whereas we interestingly found that HO-3867 suppressed p53 expression in both p53 WT U2OS and p53 MT HOS cells (Supplementary Figure S1). Whether p53 can regulate the antiapoptotic effect of HO-3867 by activating several targets, including p21, Slug, and Krüppel-like factor 4 (KLF4) in human osteosarcoma, should be extensively investigated.
To improve the bioavailability and potency of curcumin, the new synthetic curcuminoid HO-3867 was developed to target cytotoxicity toward cancer cells without influencing normal cells [18,25,26,36]. Using various experimental practices, we discovered that the addition of HO-3867 activates apoptotic processes via extrinsic and intrinsic pathways through an activation of the caspase cascade, IAPs, and the phosphorylation of MAPK pathways. Although the phosphorylation of ERK 1/2 and JNK 1/2 was observed in the study, we applied different combinations of inhibitors to confirm the exact pathway used by the analog to initiate apoptosis. Although the application of HO-3867 expectedly increased cleaved caspases 3, 8, 9, the additional application of the JNK inhibitor significantly repressed the increased values of the proteins while the co-treatment of the inhibitor of ERK did not show any suppressive effects. These findings suggest that the JNK-signaling pathway plays a critical role in the activation of apoptosis within the osteosarcoma U2OS and HOS cell lines through intrinsic and extrinsic processes, after treatment with HO-3867 but not through the ERK-signaling pathway. However, further studies are required to investigate whether the detailed results in vivo are similar to those in vitro and the positive efficacy of HO-3867 on human osteosarcoma could be obtained in clinical trials.

Conclusions
In conclusion, through our research and experiments, we investigated the anticancer efficacy of HO-3867 and investigated its mechanisms on human osteosarcoma cells. As shown by the results of the study, HO-3867 demonstrated its apoptotic mechanisms in human osteosarcoma; the combined results of HO-3867 on cancer from previous as well as the current study suggests the usefulness of HO-3867 for treating osteosarcoma.