Vitamin E TPGS 1000 Induces Apoptosis in the K562 Cell Line: Implications for Chronic Myeloid Leukemia

Chronic myeloid leukemia (CML) is a hematologic malignancy derived from the myeloid lineage molecularly characterized by t(9;22)(q34;q11) resulting in BCR-ABL1 gene fusion, which is known as Philadelphia (Ph) chromosome. Although tyrosine kinase inhibitors (TKIs) have restored and maintained the quality of life of patients with CML, an important minority of patients become resistant to first-and-second-generation TKIs and require an alternative treatment. The K562 cell (Ph+, p53-/-) line was treated with Vit E TPGS 1000 (20–80 μM) only or with other products of interest (e.g., antioxidant N-acetylcysteine (NAC), specific JNK and caspase-3 inhibitor SP600125, and NSCSI, respectively) for 24 h at 37°C. Cells were analyzed by fluorescence microscopy (FM), flow cytometry (FC), and Western blotting (WB) techniques. We show that TPGS induces apoptosis in K562 cells through H2O2 signaling mechanism comprising the activation of a minimal molecular cascade: the kinase JNK>the transcription factor c-JUN>the activation of BCL-only BH3 proapoptotic protein PUMA>loss of mitochondrial membrane potential (ΔΨm)>activation of caspase-3>chromatin condensation>fragmentation of DNA. Additionally, TPGS oxidizes the stress sensor protein DJ-1-Cys106-SH into DJ-1-Cys106-SO3 and arrested the cell cycle in the S phase. Remarkably, NAC, SP600125, and NSCSI blocked TPGS-induced OS and apoptosis in K562. Since TPGS is safe in mice and humans, it is especially promising for preclinical and clinical CML leukemia research. Our findings support the view that oxidation therapy offers an important opportunity to eliminate CML.


Introduction
Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm of the bone marrow [1] characterized by a genetic alteration known as chromosome Philadelphia (Ph), which consists of the fusion of the Abelson murine leukemia (ABL1) gene on chromosome 9 with the breakpoint cluster region (BCR) gene on chromosome 22 [2]. This gene fusion results in the expression of an oncoprotein termed BCR-ABL1 [3,4]. According to the American Cancer Society (https://www.cancer.org/), there will be 9110 new cases diagnosed with CML, and 1220 people will die of CML in 2021. CML represents about 15% of all new cases of leukemia in the United States and mainly affects adults with an average age at diagnosis around 64 years, but rarely seen in children [5]. Even though the current first-, second-, and thirdgeneration tyrosine kinase inhibitors (TKIs) are being used for treating patients with CML with great success [5,6], still some patients (~10%-15%) fail to TKI treatment, due to mutations within the kinase domain of BCR-ABL1 protein [3,7]. Therefore, it is urgent to investigate alternative therapeutic approaches such as natural product compounds and organic/synthetic new chemical entities [8][9][10][11] for CML treatment.
Vitamin E D-α-tocopheryl polyethylene glycol succinate (TPGS) is a synthetic derivative of natural α-tocopherol, prepared from the esterification of α-tocopheryl succinate (α-TOS) and polyethylene glycol (PEG) 1000 [12]. Although TPGS has FDA approvement as a safe adjuvant and is widely used in drug delivery systems as a single agent, it has proved effective as an anticancerogenic compound in human cells such as lung [13], prostate [14], breast [12,15,16], pancreatic [12], acute lymphoblastic leukemia (ALL) [17], neuroblastoma [18], and hepatocarcinoma [19] cancer cells. Although the mechanism of TGPS has not yet been fully elucidated, TPGS kills cancer cells through various mechanisms, including the inhibition of the activity of ATP-dependent Pglycoprotein overcoming multidrug resistance [12], by reactive oxygen species-(ROS-) induced apoptosis-a regulated cell death [17], cell cycle arrest, and apoptosis [16,19], through both caspase-dependent and -independent DNA damage and dominant caspase-independent programmed cell death [16]. Studies on the effect of TPGS in leukemia cells are scarce [20]. Consequently, no data are available to establish whether TPGS induces apoptosis in CML and whether the mechanism of TPGS-induced cell death is similar to a ROS-induced signaling mechanism previously demonstrated in ALL [17]. Furthermore, it is not yet known whether BCR-ABL kinase confers CML resistance to TPGS stimuli. Therefore, to get insight into these issues, we have selected the K562 cell line as a model cell [21] to investigate the cytotoxic effect of TPGS. Indeed, the K562 is a well-characterized cell line [22] that expresses the Philadelphia chromosome, i.e., BCR-ABL [23,24] responsible for maintaining proliferation, inhibiting differentiation, and conferring resistance to cell death [25]. Importantly, the K562 cell line does not express p53 [26,27]-a transcription factor implicated in cell cycle regulation, chemical-induced OS response, and apoptosis [28,29]. Therefore, it is imperative to answer whether K562 is resistant to TPGS exposure.
In this study, we determine for the first time that TPGS induces apoptosis in the K562 cell line mediated by the OS mechanism. The mechanism involves generating H 2 O 2 , the oxidation of the redox sensor DJ-1 protein into DJ-1-Cys106-SO 3 (sulfonate) derivative; activation of the proapoptotic c-JUN transcription factor; the expression of PUMA; the loss of mitochondrial membrane potential (ΔΨ m ); the activation of the protease caspase-3; and nuclear fragmentation, as apoptotic markers. We found that the BCL-2/BAX ratio was unaffected by TPGS in K562. It is also shown that the antioxidant N-acetylcysteine (NAC) and pharmacological inhibition of JNK (SP600123) and caspase-3 (NSCI) protect K562 against the cytotoxic effect of TPGS. In agreement with others, our findings support the use of TPGS as a treatment for patients with CML. Briefly, a cryovial containing the frozen K562 cells was rapidly (<1 min) thawed in a 37°C water bath. Then, cells were incubated using a prewarmed growth medium composed of RPMI 16-40 medium with L-glutamine and sodium bicarbonate (cat. # R8758, Sigma-Aldrich, St Louis, Missouri, USA), fetal bovine serum (FBS) 10% and 100 U/mL penicillin, and 100 mg/mL streptomycin. When the cells were ready for passaging (i.e., log-phase growth before they reach con-fluency), they were subcultured. Cell suspensions at passages 3 to 5 were used for further experiments. The cell suspension (1 × 10 6 cells/well in 1 mL final volume) was exposed to increasing α-tocopherol polyethylene glycol 1000 succinate (TPGS, CAS Number 9002-96-4, Sigma-Aldrich, St Louis, Missouri, USA) concentrations (10,20,40,60, and 80 μM). TPGS was prepared in PBS and stored to −20°C in the absence or presence of different products of interest (e.g., antioxidant and inhibitors) for 24 h at 37°C. The 3,30-dihexyloxacarbocyanine iodide (DiOC 6 (3)) (cat. # D-273, Thermo Fisher Scientific Inc.) and 1,9-pyrazoloanthrone (SP600125, cat. No. 420119) were purchased from Calbiochem (Merck Millipore). The dichlorofluorescein diacetate (DCFH 2 -DA) was from Invitrogen. Propidium iodide (PI) was acquired from BD Bioscience (San Jose, CA). All other reagents were from Sigma-Aldrich (St Louis, Missouri, USA).

Morphological Assessment of Cell Death by Fluorescence
Microscopy. The cell suspension (1 × 10 6 cells/well in 1 mL final volume) was exposed to increasing concentrations of TPGS for 24 h at 37°C. Fluorescence microscopy analysis was performed using a Zeiss Axiostart 50 Fluorescence Microscope equipped with a Zeiss AxioCam Cm1 (Zeiss Wohlk-Contact-Linsen, Gmb Schconkirchen, Germany). The adjustment of the images obtained was performed using the software provided by the manufacturer (ZEN 2 Core). The apoptotic indices were assessed two times in independent experiments blind to the experimenter.

Determination of DNA Fragmentation and Cell Cycle by
Flow Cytometry. DNA fragmentation and cell cycle were determined using a hypotonic solution of PI according to ref. (30). After treatment, 1 × 10 5 cells were washed with PBS (pH 7.2) and stored in 95% ethanol overnight at −20°C. Then, cells were washed and incubated in 400 μL solution containing propidium iodide (PI, 50 μg/mL), RNase A (100 μg/mL), EDTA (50 mM), and Triton X-100 (0.2%) for 30 min at 37°C. The cell suspension was analyzed for PI fluorescence by using an Epics XL flow cytometer (Beckman Coulter). Quantitative data and figures were obtained using the FlowJo 7.6.2 Data Analysis Software. Cells entering the sub-g1 phase were used as a marker of apoptosis (DNA fragmentation). For cell cycle analysis, the sub-g1 population was subtracted from the total acquired events, and the Dean Jett Fox analysis was performed (RMS < 10). The experiment was conducted three times, and 10000 events were acquired for analysis.  (30). The ΔΨ m was measured by cellular retention of DiOC 6 (3), which is selectively taken up by mitochondria and reflects the maintenance of ΔΨ m (ex. 450-490 nm, em. 515 nm). Cells were then analyzed using an Epics XL flow cytometer (Beckman Coulter). The experiment was conducted two times, and 10000 events 2 Oxidative Medicine and Cellular Longevity were acquired for analysis using the FlowJo 7.6.2 Data Analysis Software.

Evaluation of Intracellular Hydrogen Peroxide Levels by Flow Cytometry.
To determine intracellular H 2 O 2 levels, we used 2 ′ ,7 ′ -dichlorofluorescein diacetate (DCFH 2 -DA; Invitrogen) as described in ref. (17). Briefly, cells (1 × 10 5 ) exposed to increasing α-tocopherol polyethylene glycol 1000 succinate concentrations (TPGS 10, 20, 40, 60, and 80 μM) were then incubated with DCFH 2 -DA (5 μM) reagent for 30 min at 37°C in the dark. Cells were washed, and DCF fluorescence was determined using an Epics XL flow cytometer (Beckman Coulter). The assessment was repeated two times in independent experiments. Quantitative data were obtained as described above.  The pharmacological inhibitor concentration was determined in previous experimental settings in our laboratory [17]. The cells were then evaluated for DNA fragmentation and ΔΨ m by flow cytometry. The assessment was repeated two times in independent experiments.
2.9. Statistical Analysis. Statistical analyses were performed using the GraphPad Prism 6 scientific software (GraphPad, Software, Inc. La Jolla, CA, USA). Data are expressed as the mean ± SD of a minimum of two independent experiments. One-way ANOVA with a Tukey post hoc test was used to compare the differences between the experimental groups.

TPGS Generates H 2 O 2 , Oxidized the Stress Sensor
Protein DJ-1, and Activates Caspase-3 in CML K562 Cells. Next, we assessed whether TPGS produces ROS in K562 cells. Figure 3 shows that the fluorescence intensity of H 2 O 2 -3 Oxidative Medicine and Cellular Longevity sensing fluorescent probe DCFH-DA increased dosedependently by TPGS in K562 cells. To confirm the generation of ROS (specifically H 2 O 2 ) and simultaneous activation of the protein caspase-3 responsible for DNA fragmentation, we used activated caspase-3 and the oxidation of sensor-specific H 2 O 2 -reacting protein DJ-1 (e.g., DJ-1-Cys106-SO 3 ) as a probe [31] in cells treated with TPGS (10-80 μM) at 37°C for 24 h. As shown in Figure 4(a), TPGS significantly increased the expression level of protein oxDJ-1 and caspase-3 (Figure 4(b)), albeit (80 μM) had the strongest

NAC Reduces TPGS-Induced Apoptosis in CML K562
Cells. To verify that the cytotoxic effect of TPGS was related to OS, K562 cells were also cotreated with the antioxidant compound N-acetyl-L-cysteine (NAC, 1 mM) without or with TPGS (10-80 μM). As shown in Figure 5(a), NAC protected K562 cells against TPGS-induced apoptosis compared to untreated cells (Figures 5(b) and 5(c)).

TPGS Increases the Expression Level of and Activates
Caspase-3 in K562 Cells. Activation of caspase-3 has been recognized as an essential caspase for DNA fragmentation and morphological changes linked to apoptosis. This feature thus constitutes a marker of this type of cell death process. As shown in Figure 8, TPGS induced a significant increase in the expression level of caspase-3 (1.5-f.i.) according to Western blotting (Figures 8(a) and 8(b)). Accordingly, TPGS significantly activated caspase-3 (e.g.,~11%-68% CASP3 + cells) in K562 (Figure 4(a)). To confirm the involvement of caspase-3, cells were exposed to TPGS in the absence or presence of the specific inhibitor NSCI (10 μM). Flow cytometry analysis revealed that NSCI drastically reduced the apoptosis signs in treated cells with TPGS ( Figure 8(a) (Q1+Q2)) compared to untreated cells. However, similar to SP600125, the NCSI induced a significant increase in the percentage of UCD (Figures 8(d) (Q2), 6(d), and 6(e)).

Discussion
The pharmaceutical industry has used vitamin E TPGS or TPGS (also known as Tocophersolan, PubChem CID: 71406) [32,33] as a solubilizer, emulsifier, permeation, bioavailability enhancer of hydrophobic drugs, and as an excellent drug deliver agent [34]. Therefore, it has been globally accepted as a safe and nontoxic compound by the principal regulatory agencies [35]. Unsurprisingly, TPGS showed no significant toxic effects in in vitro noncancer cells (e.g., peripheral blood lymphocytes) [17] and in vivo [19]. However, TPGS has also proved an effective agent to eliminate several cancer cell lines, including ALL cells [17] and in vivo tumorigenic cells [19]. Here, we report for the first        biochemical metabolism. TPGS is a water-soluble amphipathic formulation of D-α-tocopherol succinate coupled, through a succinate linker, to polyethylene glycol (PEG) 1000; it is easily taken up into cancer cells. After hydrolysis (e.g., by cytosolic esterases), the fat-soluble D-α-tocopherol is then released either couple to acetate, forming the α-tocopheryl succinate (α-TOS) or α-tocopherol (Vit E). While Vit E is harmless to cells, mounting evidence has shown that α-

11
Oxidative Medicine and Cellular Longevity TOS induced dissipation of ΔΨ m , inhibition of mitochondrial complex I or II, and accumulation of ROS [36,37]. Whatever the mechanism, we found that TPGS significantly increased the DCF + cells reflecting the intracellular generation of ROS (H 2 O 2 ). It is well established that H 2 O 2 acts as a second messenger [38] involved in redox signaling (e.g., activation of apoptosis signal-regulating kinase 1, ASK-1) [39]. Moreover, it has been demonstrated that H 2 O 2 specifically oxidized the Cys106-SH residue of the DJ-1 protein into DJ-1-Cys106-SO 3 [31,40]. We found that TPGS-induced a   Once TPGS enters the cell, it is metabolically processed by cytoplasmic esterases and converted into alpha-tocopherol acetate (α-TOS, s2). This compound targets mitochondrial complex I or II (s3), resulting in an over generation of ROS-H 2 O 2 (s4). The signaling molecule H 2 O 2 either oxidized the oxidative sensor protein DJ-1-Cys106-SH into DJ-1-SO 3 (s5) or indirectly activated prodeath kinases (ASK-1 (s6) and MKK4 (s7)) and JNK (s8), which in turn activate c-JUN (s9). This transcription factor transcribes proapoptotic PUMA (s10), contributing to the permeabilization of the outer mitochondrial membrane (s11). Mitochondrial damage allows the release of apoptogenic proteins such as cytochrome c and ATP, which are responsible for the formation of an apoptosome complex (s12) and activation of caspase-3 protease (s13). This protease in turn activates the endonucleases DFF40/CAD, by cutting the nuclease's inhibitor DFF45/ICAD. Finally, DFF40/CAD causes nuclear chromatin fragmentation (s14), typical of apoptosis. Remarkably, the antioxidant N-acetylcysteine (NAC, red stop signs in s4), the specific JNK inhibitor SP600125 (red stop sign in s8), and the specific caspase-3 inhibitor NSCI (red stop sign in s13) block TPGS-induced apoptosis in K562 ratifying the involvement of OS signaling and caspase-3 as end-executor protein in the apoptotic pathway in this leukemia cell line. The TPGS-induced cell death mechanism provides the basis for an oxidative therapy strategy to combat leukemia.
significant elevation in the oxidized protein DJ-1. Taken together, these findings suggest that H 2 O 2 is involved in TPGS-induced apoptosis in K562. Furthermore, the antioxidant NAC completely protected K562 cells against TPGS toxic stimuli ( Figure 5). Taken together, these observations imply that TPGS produces H 2 O 2 at mitochondria, and this effect is associated with the dramatic loss of the mitochondrial membrane potential. We also determine that TPGS arrested the cell cycle in the S phase ( Figure 2). How does TPGS block the cell cycle in the S phase where the p53-a tumor suppressor gene that normally stops the progression of the cell cycle, is not expressed in K562? We speculate that TPGS can disable important proteins (e.g., cyclin B) involved in the transition from S to the G2-M phase [41]. However, further investigation is needed to clarify this issue. Previous studies have implicated JNK kinase, phosphorylated c-JUN, PUMA, and caspase-3 in TPGS-induced apoptosis in ALL cells [17]. Here, we confirmed that TPGS induces p-c-JUN and overexpresses PUMA in the K562 cell line. Furthermore, we found that the specific inhibitor JNK SP600125 completely decreased the toxic effect of TPGS according to mitochondrial ΔΨ m assay. Taken together, these results suggest that JNK kinase and c-JUN are critical molecules in the cell death process of this cell line. Indeed, once p-c-JUN is activated, this transcription factor transcribes the proapoptotic protein PUMA [42]. In agreement with this view, we found a significant increase in the expression levels of the protein PUMA according to Western blotting ( Figure 6). Since p53 is a natural transcription factor that overexpresses PUMA [43,44] and BAX [45], our observation implies that TPGS can induce apoptosis independently of p53 in K562 [46]. Consequently, we found no changes in the proapoptotic/antiapoptotic protein BAX/BCL-2 ratio [45]. This implies that PUMA is capable of lessening the ΔΨ m [47]. Likewise, Western blotting and flow cytometry analyses revealed that TPGS significantly activated caspase-3. Remarkably, the inhibitor NCSI drastically reduced caspase-3 activation in the presence of TPGS. Taken together, these results suggest that JNK/c-JUN, PUMA, caspase-3, and mitochondria are important players in TPGSinduced apoptosis in K562 cells.

Conclusion
Overall, TPGS is a promising vitamin E synthetic-derived antileukemic agent due to its prooxidant activity and specific targeting of mitochondria. Although its mechanism of action is not yet fully established, our data suggest that TPGS induced apoptosis in K562 by an H 2 O 2 -dependent but p53independent signaling mechanism. In addition to plasma membrane damage, the TPGS-induced apoptosis mechanism comprises a minimal step initially triggered by H 2 O 2 and ends up with the activation of protease caspase-3 and nucleus fragmentation (Figure 9). Since TPGS is safe in mice (e.g., a dose of 100 mg/kg through tail-vein injections) [19], TPGS is especially promising for preclinical leukemia research. Considering the present results, detailed biological studies are justified to determine conditions under which the proox-idant properties of TPGS serve to actively burst OS in CML cells.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Disclosure
The funder had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.