Role of 2′-Hydroxycinnamaldehyde In The Induction of Apoptosis Via A Reactive Oxygen Species-Dependent JNK Pathway in Human Promyelocytic HL-60 Leukemia Cells

Background: Because the role and molecular mechanism of 2′-hydroxycinnamaldehyde (2′-HCA) in human leukemia need to be claried, we provide detailed insights into the mechanism underlying the anti-proliferative properties of 2′-HCA in acute myeloid leukemia (AML). Methods: We performed MTT assay to examine the cytotoxic effect of 2′-HCA. 2′-HCA-induced apoptotic pathway was demonstrated by annexin V-FITC/ propidium iodide double staining, DAPI staining for DNA fragmentation detection, and western blot analysis. ROS generation, intracellular glutathione (GSH) level, and intracellular protein thiols (PSH) were detected to investigated the mechanism of 2′-HCA-related apoptosis. Xenograft animal model was used to evaluate anti-tumor activities of 2′-HCA. Results: The present study demonstrated that 2′-HCA induced apoptosis in human promyelocytic leukemia HL-60 cells through the activation of mitochondrial pathways. 2′-HCA also induced the activation of JNK and the pharmacological inhibition of JNK effectively prevented 2′-HCA-induced apoptosis and AP-1-DNA binding. In addition, 2′-HCA resulted in the accumulation of ROS and depletion of intracellular GSH and PSH in HL-60 cells. Xenograft mice inoculated with HL-60 leukemia cells demonstrated that the intraperitoneal administration of 2′-HCA inhibited tumor growth by increasing of TUNEL staining, the expression levels of nitrotyrosine, pro-apoptotic proteins, and PCNA protein expression. Conclusion: Our ndings suggest that 2′-HCA induces apoptosis via the ROS-dependent JNK pathway and could be considered as a agent for


Background
Changes in cellular oxidative stress have emerged as a critical event in cancer. Intracellular reactive oxygen species (ROS), which are produced continuously by the mitochondria, have been suggested to regulate the processes involved in cancer cell cycle arrest, senescence, and apoptosis [1]. This can be achieved by reducing the levels of cellular antioxidants such as glutathione (GSH) and ROS including superoxide anion, nitric oxide, peroxynitrite, and hydrogen peroxide [2]. Elevated levels of ROS activate cellular signaling pathways such as mitogen-activated protein kinase (MAPK), nuclear factor kappa-B (NF-κB), Wnt, and Kelch-like ECH-associated protein 1 (Keap1)-nuclear factor (erythroid-derived 2)-like 2 (Nrf2), which can lead to apoptosis [3]. This may be attributed to an increase in mitochondrial oxidative stress that causes cytochrome c release into the cytosol, leading to caspase activation. Several anticancer and chemopreventive agents generate ROS, which are associated with the apoptotic cell death of tumor cells [2].
Leukemia is a major hematological malignancy that causes mortality and morbidity in different age groups [4]. It is de ned as the abnormal proliferation, clonality, and differentiation of immature hematopoietic cells in the bone marrow [5]. Leukemia can be classi ed into four common types: acute lymphocytic leukemia (ALL), chronic lymphocytic leukemia (CLL), acute myeloid leukemia (AML), and chronic myelogenous leukemia (CML). Among these types of leukemia, AML is characterized by the aggressive growth of hematopoietic precursor cells that interfere with the production of normal hematopoietic cells in the bone marrow [6]. Although re nement of supportive treatment has improved the outlook of patients with AML in the past 30 years, more than half of young adults and around 90% of older patients still die from AML [5]. The majority of AML cells express varying amounts of the transmembrane surface glycoprotein CD33 (observed in approximately > 80% of patients with AML). Therefore, the FDA has granted accelerated approval to gemtuzumab ozogamicin (GO), which is a humanized monoclonal antibody that binds with the IgV domain of CD33 for older patients with relapsed CD33-positive AML [7]. In addition, CPX-351 (a dual drug liposomal encapsulation of cytarabine and daunorubicin), enasidenib (a selective oral inhibitor of the mIDH2 enzyme), and midostaurin (a multitargeted kinase inhibitor active in AML patients with an FLT3 mutation) were approved by the FDA in 2017. Although the past few years have been an active period for the clinical testing and FDA approval of various molecularly targeted treatments using novel agents for AML, there is limited information on side effects including hepatotoxicity, cardiotoxicity, hematotoxicity, and infection [8].
2′-Hydroxycinnamaldehyde (2′-HCA, Fig. 1), which is an active compound isolated from Cinnamomum cassia, is a cinnamaldehyde derivative [9]. 2′-HCA is known to have anti-tumor effects in various cancer cells, which include the prevention of cell proliferation and induction of apoptosis [10][11][12]. In addition, 2′-HCA has been reported to exhibit various biological activities, including the suppression of β-catenin signaling and epithelial-mesenchymal transition (EMT) in cancer cells [13]. In addition, 2′-HCA has been found to suppress cancer cell proliferation and tumor growth via the activation of pyruvate kinase M2 [14] and signal transducer and activator of transcription 3 (STAT-3) by regulating ERK1/2 and ROS generation in prostate cancer cells [15]. Nevertheless, the role and molecular mechanism of 2′-HCA in human leukemia need to be clari ed. In the present study, we provide detailed insights into the mechanism underlying the anti-proliferative properties of 2′-HCA in AML.
MTT assay Cytotoxicity was measured using the MTT assay. The MTT assay was performed for cytotoxicity measurement using a modi ed method described [17]. Brie y, the cells (5 × 10 4 ) were seeded in each well containing 100 ml of the medium in 96-well plate. After incubation for 24 h, various concentrations of 2′-HCA were added to the 96-well plate. After 48 h, 50 ml of MTT (5 mg/ml stock solution in PBS) was added to each well for 4 h. The medium was discarded and the formazan blue which formed in the cells was dissolved with 100 mL DMSO. The optical density was measured at 540 nm.
Detection of DNA fragmentation DNA fragmentation was quantitated with DAPI assay as previously reported [18]. In brief, cells were lysed in a solution containing 5 mM Tris-HCl (pH 7.4), 1 mM EDTA, and 0.5% (w/v) Triton X-100 for 20 min on ice. After centrifugation at 27,000 ×g for 20 min, the lysate and supernatant were sonicated for 40 s, and the level of DNA was measured by a uorometric method using DAPI. The amount of the fragmented DNA was calculated as the ratio of the amount of DNA in the supernatant to that in the lysate. The genomic DNA was prepared for gel electrophoresis as previously described [19]. Electrophoresis was performed in a 1.5% (w/v) agarose gel in 40 mM Tris-acetate buffer (pH 7.4) at 50 V for 1 h. The fragmented DNA was visualized by staining with ethidium bromide after electrophoresis.
Preparation of mitochondrial and cytosolic fraction and total protein HL-60 cells (2.5 × 10 7 ) were collected by centrifugation at 200 × g for 10 min at 4°C. The cells were washed twice with ice-cold PBS, followed by centrifugation at 200 × g for 5 min. The cell pellet was then resuspended in cell lysis buffer (20 mM HEPES-KOH, pH 7.5, 10 mM KCl, 1.5 mM MgCl 2 , 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 100 mM PMSF) for 30 min on ice. Cells were then homogenized with a glass dounce and a B-type pestle (80 strokes). Cell homogenates were spun at 15,000 × g for 15 min at 4°C and the supernatant (cytosolic fraction) was removed while taking care to avoid the pellet. The resulting pellet was resuspended in mitochondrial buffer. For total cell protein extracts, cells were washed with ice-cold PBS and extracted in protein lysis buffer (Intron, Seoul, Korea) for 30 min on ice. Cells were then spun at 15,000 ×g for 30 min at 4°C and the supernatant (total protein) was used to detection for protein expression.

Western blot analysis
For western blot analysis, protein concentration was determined by Bradford assay. Protein samples were mixed with 5 × SDS sample buffer, boiled for 4 min, and then separated by SDS-PAGE gels. After electrophoresis, proteins were transferred to polyvinylidene di uoride membrane. The membranes were incubated with the primary antibody in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) overnight at 4°C. The primary antibody of the membrane was removed by washing in TBS-T and the membrane was incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h. Following washing in TBS-T, immune blots were visualized by ECL (GE Healthcare) and exposed to X-ray lm (Amersham, NJ, USA).

Determination of mitochondrial membrane potential (DΨ m )
Changes in the DΨ m were examined by monitoring the cells after double staining with PI and rhodamine 123. After treatment of 40 mM 2′-HCA for 2 h, the cells were incubated with medium containing 5 mg/mL rhodamine 123 for 1 h to determine the mitochondrial membrane potential. The cells were resuspended in 1 mL of minimal essential medium containing 5 mg of PI to assess cell viability. The intensity of uorescence from PI and rhodamine 123 was measured by ow cytometry. Fluorescence was measured after the cells staining for 30 min at 37 °C.

Detection of ROS generation
To measure 2′-HCA-induced intracellular ROS level, we used DCFH-DA, which is the most widely used uorescent probe for the detection of intracellular oxidative stress [20]. The 2′-HCA-treated cells were incubated with 20 mM DCFH-DA for 30 min at 37 o C. The intracellular ROS level was measured by ow cytometry.

Determination of the GSH level
Cells were washed twice with PBS and treated with 5% trichloroacetic acid (TCA) to extract cellular GSH. The mixture was centrifuged at 13,000 × g for 1 min to remove the denatured proteins. GSH was determined by the enzymatic method as previously described [21]. To determine the glutathione disul de (GSSG), the same DTNB recycling assay was performed after using 2-vinylpyridine to remove the reduced GSH [22]. Brie y, 2 μL of 2-vinylpyridine and 6 μL of triethanolamine were simultaneously mixed with 100 μL of sample, followed by incubation in the dark at room temperature for 1 h before initiation of the recycling assay. The kinetics of the reaction was monitored for 10 min. The increment in absorbance at 412 nm was converted to GSH concentration using a standard curve with known amounts of GSH.

Measurement of intracellular protein thiols (PSH)
To measure the intracellular PSH, we performed the assay of intracellular PSH as previously report [22]. Brie y, cells were treated with 5% TCA and then vortexed and kept on ice for 30 min to prepare complete protein precipitation. After centrifugation, the protein precipitate dissolved in 0.1 M Tris-HCl buffer (pH 7.5), containing 5 mM EDTA and 0.5% sodium dodecyl sulfate (SDS). One aliquot of this protein precipitate was reacted with a solution containing 0.1 M sodium phosphate buffer (pH 7.5), 5 μM EDTA, 0.6 mM DTNB, 0.2 mM NADPH, 1 unit/ml glutathione reductase, and another aliquot of the protein solution was treated with 5 mM N-ethylmaleimide (NEM) before the reaction to obtain the background value for subtraction. The concentration of intracellular protein thiol was expressed as nmol of SH equivalents/mg protein using GSH as a standard.
Caspase-3 activity assay The caspase-3 activity was measured using a uorogenic caspase-3 substrate (Ac-DEVD-AFC). Cells were washed once with PBS, resuspended in 400 μl of lysis buffer (20 mM HEPES, pH 7.4, 100 mM NaCl, 0.5% NP-40, 10 mM DTT) and incubated on ice for 30 min. After centrifugation (12,000 × g for 5 min), supernatants were collected and immediately measured for protein concentration and caspase activity, or stored at -70°C until assayed. For the activity assay, 100 μL of cell lysates were placed in a 96-well plate and a caspase substrate was added to each well. Plates were incubated at 37 °C for 1 h and caspase activity was determined from the uorescence read at 505 nm induced by excitation at 400 nm.
Transfection for RNA interference HL-60 cells (5 × 10 6 cells) were collected by centrifugation at 200 ×g for 10 min at 4 ºC. The cells were washed twice with ice-cold PBS, pH 7.2, followed by centrifugation at 200 ×g for 5 min. Nuclear extracts and EMSA assay were performed as described previously [23].

Animals
The male BALB/c nude mice (6-week-old, 20-23 g) were obtained from Nara Biotec Co. (Pyeongtaek, Republic of Korea). Mice were inhabited 6/cage/group and were had standard laboratory chow in an animal room with 12 h dark/light cycles at a constant temperature of 20 ± 5 ºC. All animal experiments were performed under university guidelines and were approved by the ethical committee for Animal Care and Use of Kyung Hee University according to the animal protocol (KHUASP(SE)-19-027).

Xenograft animal model
The male BALB/c nude mice were subjected to 150 mg/kg cyclophosphamide (CYP) by intraperitoneal (i.p.) injection three times per two days (Fig. 2). After CYP injection, HL-60 cells (1 × 10 6 per site) were inoculated subcutaneously into the right side of the ank of male BALB/c nude mice. Tumor size was checked with a caliper [24] once per 3 days in a week and calculated as V = π/6 × (length) × (width) 2 .
When tumor volume reached around 300 mm 3

Statistical analysis
All data presented as means ± SD were analyzed by using GraphPad Prism 8.0 Software (San Diego, CA, USA). Student's t-test analysis was applied for statistical analysis to compare all the different groups in the current study. The difference was considered to have statistical signi cance if P < 0.05.

2′-HCA induces apoptosis in HL-60 cells
Initially, we performed MTT assay to examine the cytotoxic effect of 2′-HCA on various cancer cells (Table  1). Among the tested cancer cell lines, HL-60 and Molt-4 were more susceptible to 2′-HCA. Subsequently, we examined whether the cytotoxicity of 2′-HCA is attributed to apoptotic cell death. As shown in Fig. 3A, Annexin V-positive cells (early apoptotic cells) were increased in a time-dependent manner after the treatment of the HL-60, Molt-4, U937, and K562 leukemia cells with 2′-HCA. In agreement with the results of MTT assay, 2′-HCA-induced apoptosis was most pronounced in HL-60 cells. Therefore, we selected HL-60 cells to further investigate the apoptotic mechanism. As shown in Fig. 3B and 3C, 2′-HCA increased the quanti cation and laddering pattern of internucleosomal DNA fragmentation in HL-60 cells. These results indicated that 2′-HCA-induced leukemia cell death was caused by apoptosis and that HL-60 cells were highly reactive with 2′-HCA.

2′-HCA-induced apoptosis is involved in mitochondrial dysfunction and caspase activations in HL-60 cells
The process of cell death may involve mitochondrial dysfunction including the release of cytochrome c from the mitochondria by inducing the oligomerization of Bcl-2 family proteins, which subsequently causes apoptosis via caspase activation [25]. Treatment with 10 μM 2′-HCA time-dependently increased the translocation of Bim and Bax from the cytosol to mitochondria and cytochrome c release from the mitochondria into the cytosol, and cellular Bcl-2 protein levels were reduced (Fig. 4A). As cytochrome c release is caused by ΔΨ m disruption, we next evaluated the effect of 2′-HCA on ΔΨ m by ow cytometry after double staining with Rh123 and PI (Fig. 4B). The results showed that control cells appeared mostly on the Rh123 high-uorescence (+) PI (−) (lower right quadrant) eld, whereas 2′-HCA-treated HL-60 cells showed an increasing cell population on the Rh123 low-uorescence (−) PI (−) (lower left quadrant) eld. Furthermore, we examined the time-dependent proteolytic cleavage of procaspase-9, procaspase-3, and PARP-1 in HL-60 cells; however, the cleavage of procaspase-8 was not detected (Fig. 4C). To determine whether caspase activation is required for 2′-HCA-induced apoptosis, we pretreated HL-60 cells with caspase inhibitors. As shown in Fig. 4D, z-VAD-FMK (a broad caspase inhibitor) and z-DEVD-FMK (a caspase-3 inhibitor) inhibited 2′-HCA-induced DNA fragmentation, whereas Ac-IETD-CHO (a caspase-8 inhibitor) did not affect 2′-HCA-induced apoptosis. These observations indicated that the caspasedependent mitochondrial intrinsic pathway could be involved in 2′-HCA-induced apoptosis.
2′-HCA-induced apoptosis is regulated by JNK activation in HL-60 cells Among the various signaling pathways that respond to stress, the mitogen-activated protein kinase (MAPK) signaling pathway is crucial for apoptosis [26]. To investigate signal transduction events that could contribute to apoptosis, we determined the role of the MAPK pathway in 2′-HCA-treated HL-60 cells.
We found that the exposure of HL-60 cells to 2′-HCA resulted in a time-dependent increase in the phosphorylation of apoptosis signal-regulating kinase-1 (ASK-1), JNK, and ERK1/2 but not p38 MAPK (Fig. 5A). ASK-1, JNK, and ERK1/2 activation was evident as early as 15 min after treatment with 10 µM 2′-HCA and persisted for the duration of the experiment in HL-60 cells. Commercially available p-JNK inhibitor (SP600125), ERK1/2 inhibitor (U0126), and p38 MAPK inhibitor (SB203580) were used to further determine whether 2′-HCA exerts its sensitizing effects by inhibiting p-JNK, p-ERK1/2, and p-38 MAPK. Interestingly, among the MAPK inhibitors, SP600125 pretreatment signi cantly inhibited 2′-HCA-induced DNA fragmentation, whereas there was no protective effect by either U0126 (ERK1/2 inhibitor) or SB203580 (p38 MAPK inhibitor) pretreatment up to 8 h (Fig. 5B). To rule out the possibility of the nonspeci c effect of 2′-HCA on JNK, we analyzed the effect of JNK silencing. As shown in Fig. 5C and 5D, the knockdown of JNK by siRNA markedly blocked 2′-HCA-induced PARP-1 cleavage and DNA fragmentation in HL-60 cells, which were similar to the effects observed following treatment with the chemical inhibitor SP600125. JNK modulates the apoptotic pathway including the activation of speci c transcription factors [27]. We evaluated JNK activity and AP-1 DNA-binding activity with JNK kinase assay and EMSA, respectively, in 2′-HCA-treated HL-60 cells. As shown in Fig. 5E and 5F, JNK and AP-1 DNA-binding activities were signi cantly increased by treatment with 10 μM 2′-HCA within 2 h, which was earlier than the onset of apoptosis as detected by caspase-3 activation and apoptotic DNA fragmentation. These observations indicated that the JNK pathway could play a crucial role in 2-HCAinduced apoptosis via the regulation of transcription factors.
Oxidative stress is required for 2′-HCA-induced apoptosis in HL-60 cells ROS have been demonstrated to be an early signal that mediates apoptosis [28]. Indeed, it has been reported that 2′-HCA could induce the apoptosis of cancer cells mainly through ROS generation [11,15].
To con rm whether ROS are involved in 2′-HCA-mediated apoptosis in leukemia cells, we measured the levels of cellular ROS using DCFH-DA with a uorescence microscope. As shown in Fig. 6A, marked ROS generation was observed in 5min after treatment with 10 μM 2′-HCA, and the 2′-HCA-induced ROS generation was signi cantly reduced with antioxidant N-acetylcysteine (NAC) pretreatment (Fig. 6B). As increasing evidence has suggested that the intracellular thiol redox status is one of the key mediators of apoptosis in many cell systems [29], we examined whether 2′-HCA-induced apoptosis involves the depletion of intracellular thiols. 2′-HCA rapidly reduced the levels of intracellular GSH and PSH in a timeand concentration-dependent manners and a statistically signi cant difference was detected as early as 15 min after treatment with various concentrations (2.5, 5, or 10 μM) of 2′-HCA ( Fig. 6C and 6D). To examine whether the generation of ROS is a crucial step in 2′-HCA-induced apoptosis, we investigated the effect of NAC on 2′-HCA-induced apoptosis. As shown in Fig. 6E and 6F, pretreatment with NAC decreased the 2′-HCA-induced sub-G 1 cell population and abrogated the translocation of Bim and Bax from the cytosol to mitochondria, reduction of Bcl-2 protein expression, and release of cytochrome c into the cytosol. In addition, pretreatment with NAC signi cantly attenuated 2′-HCA-induced caspase-3 activity in HL-60 cells (Fig. 6G). These results demonstrated that oxidative stress could play an important role in 2′-HCA-induced apoptosis in HL-60 cells.
Reduced oxidative stress attenuates the 2′-HCA-induced JNK pathway and mitochondrial translocation of Bim in HL-60 cells ROS can promote the activation of JNK, which modulates the activity of the proapoptotic BH3 subgroup of Bcl-2 family proteins such as Bim [1]. As 2′-HCA activated the JNK pathway and AP-1 transcription factor activity, we investigated whether the JNK pathway is associated with oxidative stress in 2′-HCAtreated HL-60 cells. As shown in Fig. 7A, pretreatment with NAC effectively attenuated 2′-HCA-induced ASK-1 protein expression and JNK phosphorylation but did not affect the phosphorylation of ERK1/2 and p38 MAPK. In addition, pretreatment with NAC completely inhibited 2′-HCA-induced AP-1 DNA-binding activity, indicating oxidative stress-regulated JNK/AP-1 signaling in HL-60 cells (Fig. 7B). Furthermore, our results revealed that the mitochondrial translocation of Bim was blocked by treatment with NAC and SP600125, indicating the important role of Bim in the 2′-HCA-activated ROS/JNK pathway (Fig. 7C). These results demonstrated that 2′-HCA-induced apoptosis could mediate mitochondrial dysfunction through mitochondrial Bim translocation regulated by the ROS/JNK pathway in HL-60 cells.
2′-HCA suppresses tumor growth in a HL-60 xenograft mouse model To evaluate the anti-tumor effect of 2′-HCA, we made a subcutaneous xenograft model of HL-60 cells in immunode cient mice [30]. As shown in Fig. 8A, the average tumor volume was similar in each group at the start of the experiment. After intraperitoneal administration of 2′-HCA, tumor growth was slower than that of the vehicle-treated control group, and the tumor volume was markedly decreased from day 10. In addition, on the last day of the experiment, the tumor volume was signi cantly smaller in the 2′-HCA administration group (20 mg/kg, i.p.) than in the control group (2525.77 ± 1316.10 mm 3 vs. 780.47 ± 665.28 mm 3 , P < 0.001). Consistently, IHC results showed the marked reduction of the proliferating cell nuclear antigen (PCNA) marker of proliferating cells in the tumor tissues of 2′-HCA-treated mice (Fig. 8C). Similar to in vitro results, 2′-HCA treatment increased the expression levels of nitrotyrosine, a marker of ROS, and enhanced apoptosis induction, as demonstrated by TUNEL staining of tumor tissues. Furthermore, western blotting revealed that treatment with 2′-HCA increased the levels of ASK-1, p-JNK, and the pro-apoptotic mediator Bim, resulting in the cleavage of PARP-1 (Fig. 8D). During the experimental period, 2′-HCA administration did not affect the body weight, and 2′-HCA showed no toxicity in the HL-60 xenograft mouse model (Fig. S1). Taken together, our ndings indicated that 2′-HCA could inhibit tumor growth in vivo via ROS generation and JNK activation, which was consistent with the in vitro nding showing apoptosis induction via the ROS-dependent JNK pathway.

Discussion
ROS play a vital role in various cellular processes under physiological and pathological conditions. Excessive cellular levels of ROS can trigger oxidative stress, which is de ned as a severe redox imbalance between the generation of ROS and antioxidant defenses, causing oxidative damage [31]. A moderate increase in ROS can promote cell differentiation and proliferation; however, excessive ROS accumulation causes oxidative stress and damage to cells [32]. Therefore, the anti-tumorigenic signaling of ROS may be targeted in cancer therapy by increasing the production of ROS to toxic levels and exhausting the antioxidant system [33]. Indeed, several studies have reported promising compounds that could elicit ROS generation, leading to apoptosis [34,35]. In our previous studies, we found that costunolide induced apoptosis in human ovarian cancer and leukemia cells via ROS generation and JNK activation [22,36,37], and cinnamaldehyde induced apoptosis via ROS-mediated mitochondrial permeability transition in HL-60 cells [38]. 2′-HCA, which is a derivative of cinnamaldehyde, also induced ROS-mediated apoptosis with STAT-3 activation, which was abrogated by GSH or NAC treatment in DU145 cells [15]. Although 2′-HCA can inhibit the growth of human erythroleukemia or skin cancer cells by directly targeting Pim-1 kinase [39], limited studies have been conducted on the ROS-related molecular mechanism of 2′-HCA in human leukemia cells. Therefore, in the present study, we investigated the ROS-mediated anticancer mechanism of 2′-HCA in HL-60 cells. Interestingly, compared with our previous ndings, the results here indicated that 2′-HCA (IC 50 = 4.17 μM) was more cytotoxic than cinnamaldehyde (IC 50 = 30.7 μM) in HL-60 cells. It is known that the aldehyde group of the side chain and free hydroxy-substituted groups play a critical role in the anti-tumor activity of cinnamaldehydes [40]. Therefore, we believe that the increased cytotoxicity of 2′-HCA may be attributed to a difference in the structure of the H group at the 2′-site of cinnamaldehyde, which is displaced by the hydroxyl group. 2′-HCA induced intracellular ROS generation in HL-60 cells, as detected by the ROS-sensitive uorescent dye DCFH-DA, which is used as a probe for the speci c detection of intracellular hydrogen peroxide rather than superoxide radicals. Similar to our in vitro results, 2′-HCA increased the expression of nitrotyrosine, which can be targeted to elevate cellular ROS levels in tumor tissues. Moreover, we found that pretreatment with NAC abrogated mitochondriadependent apoptosis including the translocation of Bim and Bax from the cytosol to mitochondria and the release of cytochrome c into the cytosol in 2′-HCA-treated HL-60 cells. These results suggest the involvement of oxidative stress in 2′-HCA-induced intrinsic mitochondrial apoptosis in HL-60 cells.
ROS have been implicated in the activation of various cellular signaling pathways including MAPK, which can activate cell survival and/or cell death processes such as apoptosis [41]. ASK-1, a member of the MAPK kinase kinase (MAPKKK or MAP3K) family, is part of the MAPK cascade and binds to reduced thioredoxin in non-stressed cells. Under oxidative stress, thioredoxin is oxidized and dissociates from ASK-1, leading to the activation of MAPK pathways, which can promote apoptosis [42,43]. Consistently, 2′-HCA induced a marked increase in ASK-1 protein expression and the phosphorylation of ERK1/2 and JNK in this study. However, the SP600125 JNK inhibitor but not the U0126 ERK1/2 inhibitor signi cantly Experimental procedures involving the use of mice and care were conducted in accordance with institutional guidelines and compliance.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information les.

Competing interests
The authors declare no competing interests.  Figure 1 Chemical structure of 2′-HCA.

Figure 7
Effect of 2′-HCA-activated ROS on MAPK activation in HL-60 cells. (A) Representative western blots showing changes in the protein levels of ASK-1, p-ERK1/2, p-JNK, p-p38 MAPK, ERK1/2, JNK, and p38 MAPK after pretreatment with 5 mM NAC for 1 h followed by treatment with 10 μM 2′-HCA for 8 h. β-actin was used as an internal control. (B, C) After pretreatment with 5 mM NAC or 20 μM SP600125 for 1 h, HL-60 cells were treated with 10 μM 2′-HCA for 8 h, and AP-1 DNA-binding activity and Bim translocation were analyzed by EMSA and western blotting, respectively. A competition experiment using a 5-fold excess of cold oligonucleotides (C.P.) indicated that DNA binding was speci c. α-tubulin and COX 4 were used as internal controls.