TPEN Induces Apoptosis Independently of Zinc Chelator Activity in a Model of Acute Lymphoblastic Leukemia and Ex Vivo Acute Leukemia Cells through Oxidative Stress and Mitochondria Caspase-3- and AIF-Dependent Pathways

Acute lymphoblastic leukemia is still an incurable disease with resistance to therapy developing in the majority of patients. We investigated the effect of TPEN, an intracellular zinc chelator, in Jurkat and in ex vivo acute lymphoblastic leukemia (ALL) cells resistant to chemotherapy. Changes of nuclei morphology, reactive oxygen species generation, presence of hypodiploid cells, phosphatidylserine translocation, mitochondrial membrane depolarization, immunohistochemical identification of cell death signalling molecules, and pharmacological inhibition were assayed to detect the apoptotic cell death pathways. We found that TPEN induces apoptosis in both types of cells by a molecular oxidative stress pathway involving O2•− > H2O2 ≫ NF-κB (JNK/c-Jun) >p53> loss ΔΨm> caspase-3, AIF > chromatin condensation/DNA fragmentation. Interestingly, TPEN induced apoptosis independently of glucose; leukemic cells are therefore devoid of survival capacity by metabolic resistance to treatment. Most importantly, TPEN cytotoxic effect can eventually be regulated by the antioxidant N-acetyl-cysteine and zinc ions. Our data suggest that TPEN can be used as a potential therapeutic prooxidant agent against refractory leukemia. These data contribute to understanding the importance of oxidative stress in the treatment of ALL.


Introduction
Leukemia is a malignancy of hematopoietic cell populations responsible of at least 37,520 new cases and 15,200 people deaths in the United States [1]. Although the cause of leukemia is still unknown, several cellular, genetic, and biochemical alterations are the most probable mechanisms of cause [2][3][4][5]. Apoptosis is a controlled and regulated form of programmed cell death defined by specific morphological and biochemical features [6][7][8]. Reactivation of the apoptotic cell death appears as a major goal to eliminate cancer cells [9,10]. Unfortunately, secondary therapy-related leukemia (e.g., acute lymphocytic leukemia, ALL) might emerge following chemotherapy, radiotherapy, and/or terminal differentiation for primary malignancies [11][12][13]. Consequently, leukemia is still an incurable disease with resistance to therapy developing in the majority of patients. Therefore, it is necessary to investigate therapies to either achieve maximal cancer cell death or cell terminal differentiation. Given the complexity of death/differentiation pathways within a cell [8,[14][15][16], placing these pathways in the proper relationship to the drug trigger is challenging. Since metal dysregulation (e.g., zinc) has been shown in patients with leukemia [17,18] and it is essential cofactor for many proteins and transcription factors [19,20], it is reasonable to think that the use of chelators might be a potential class of pharmaceutical agents to battle different types of cancer [21,22]. Therefore, zinc depletion [19,20] in ALL patients might be a realistic goal. TPEN (N,N,N,N-Tetrakis(2-pyridylmethyl)-ethylenediamine) is a lipid-soluble zinc metal chelator and has been 2 Oxidative Medicine and Cellular Longevity shown to induce apoptosis in several cancer cells [23][24][25][26][27] through caspase family of proteases [25,[28][29][30][31][32][33], transcription factor p53 [28,33], and X-linked inhibitor of apoptosis protein (XIAP [27]). Yet, the complete molecular mechanism of cell death signaling induced by TPEN in a single cell has not yet fully established. Interestingly, the antioxidant Nacetyl-L-cysteine (NAC) inhibited the cytotoxic effect of TPEN, indicating that oxidative stress (OS) is the likely mediator of Zn-deficiency-related cell death [25]. Remarkably, molecules that generated OS induce a minimal completeness of cell death signaling pathway as a mechanistic explanation of cancer cell demise [34]. We hypothesize that TPEN might induce apoptosis by OS in leukemia cells.
To test this assumption, we sought (i) to determine whether TPEN treatment induces OS through H 2 O 2 , and activation of the proapoptotic transcription factors, kinases, and caspase-3 in Jurkat cells, used as a model of human ALL. We also wanted (ii) to determine the role played by the apoptosis-inducing factor (AIF) in TPEN treatment. Glucose appears to be critical pathophysiology resistance against OS in cancer cells [35,36]. We evaluated therefore whether glucose might alter the survival response to TPEN in Jurkat cells. To validate this in vitro data, ALL cells from one patient resistant to chemotherapy and radiotherapy were challenged with TPEN. ALL cells were similarly evaluated for nuclei morphological changes, loss ΔΨ m , p53, and caspase-3 activation or AIF analysis. Understanding the mechanism of OS may provide insight into more effective anticancer therapy.

Morphological Assessment of Cell Death by Fluorescence
Microscopy and Flow Cytometry Analysis. The cell suspension (1 mL, final volume) was exposed to increasing TPEN (0.1-5 μM) concentrations freshly prepared in RPMI-1640 medium either with glucose (G11, G55; Gibco/Invitrogen) or absence (G0) in the absence or presence different products of interest for 24 h at 37 • C. Fluorescent microscopy analysis and quantification of apoptotic morphology was performed according to [37]. The apoptotic indexes were assessed 3 times in independent experiments blind to experimenter. For annexin V/7-AAD flow cytometry analysis, cells were evaluated according to supplier's protocol (BD Pharmingen, San Diego, CA) with a flow cytometer FACSCanto II, Becton Dickinson (San José, CA). The flow cytometry apoptosis was assessed 3 times in independent experiments.

Determination of DNA Fragmentation by Flow
Cytometry. DNA fragmentation was determined by using a hypotonic solution of propidium iodide (PI). Cells entering the sub-G1 phase were used as the marker for apoptosis. Cell suspensions were analyzed in a FACScanto III flow cytometer (Beckton Dickinson). 20,000 events were assessed. Determination of DNA fragmentation was assessed 3 times in independent experiments.

Evaluation of Intracellular Reactive Oxygen Species (ROS). Superoxide anion radical (O 2
•− ) was examined for formation of formazan positive cells and quantification was evaluated according to [38]. The assessment was repeated 3 times in independent experiments.

Analysis of Mitochondrial
Membrane Potential ( ΔΨ m ) by Flow Cytometry. Jurkat cell line was treated as described above. Then, cells (1 × 10 5 ) were incubated for 20 min at room temperature in the dark with cationic lipophilic DiOC 6 (3) (10 nM, final concentration) and intercalating agent propidium iodide (PI, 12.5 ng/mL, final concentration) according to standard protocol. Cells were analyzed using a flow cytometer FACSCanto II, Becton Dickinson (San José, CA). The assessment was repeated 3 times in independent experiments.

TPEN Induces Apoptosis in Jurkat Cells Independent of Glucose Concentration but Its Toxic Effect Is Diminished by
N-Acetyl-Cysteine. As depicted in Figure 5, TPEN induces apoptosis ( Figure 5(a)), loss of ΔΨ m ( Figure 5(b)), and DNA fragmentation ( Figure 5(c)) in Jurkat cells in a glucoseand time-independent manner compared to cells cultured in glucose alone. In contrast, antioxidant compounds such as NAC (1 mM) and zinc ions (10 μM) clearly protects (Table 1) and rescues (Table 2) TPEN-induced apoptosis effect in Jurkat cells. To ascertain that TPEN induces apoptosis meanly via OS, cells were exposed to Zn(SO 4 )/TPEN complex. No significant difference in nuclei morphology, plasma membrane damage, and loss ΔΨ m were detected in Jurkat cells treated with either the Zn/TPEN complex or treated with Zn plus TPEN (Tables 1 and 2). Furthermore, no significant difference in NBT + (e.g., 17±3%) and DCF + (e.g., 13 ± 3%) percentages were observed between cells treated with Zn/TPEN complex and untreated ( Figure 2) or zinc alone (e.g., 3 ± 1% NBT + , 12 ± 2% DCF + ) for 24 h, as indication of reduced O 2 •− and H 2 O 2 production. sample with TPEN displayed nuclei changes typical of apoptosis compared to untreated cells (Figure 6(a) and inset). Noticeably, these morphologies are comparable to those displayed by Jurkat cells treated with TPEN (Figure 1(b)). Likewise, TPEN induced mitochondrial depolarization (Figure 6(g) versus Figure 6(h)). Interestingly, TPEN induces both phenomena in a concentration dependent fashion (Figure 6(i)). As shown in Figure 7, TPEN significantly induces activation of p53 (Figure 7(b)), caspase-3 (Figure 7(c)), and translocation of AIF to nuclei (Figure 7(d)) compared to cells in absence of zinc chelator (Figure 7(a)) similar to TPEN treated Jurkat cells. Clearly, not only the amount of DAB + nuclei in ALL cells treated with TPEN was significantly different to DAB + nuclei in untreated ALL cells, but also the number of AO/EB/Hoechst cells and ΔΨ m markers were significantly different compared to untreated cells (Table 3).

Discussion
In the present investigation, we report for the first time in vitro evidence of a cause and effect mechanism of apoptosis induced by the zinc chelator TPEN in Jurkat cells, and in ALL cells from a leukemic patient. The TPEN-induced apoptosis complies with the model of minimal completeness of cell death signaling induced by OS [34]. Most importantly, we demonstrated that TPEN induced cell death independently of cellular energy requirement (i.e., glucose [39]) in the culture milieu. The model provides therefore a mechanistic  explanation of leukemia cell demise. Specifically, we showed that TPEN induces apoptosis primarily via OS. We found that NAC not only significantly protect but also rescue Jurkat cells against TPEN exposure [25]. Moreover, when cells were exposed to Zn/TPEN complex, comparable O 2 •− /H 2 O 2 values to untreated cell were detected; consequently, TPENinduced apoptosis was significantly reduced. This data implies that zinc disabled TPEN to generate ROS, thereby turning down cell signalling. This observation therefore suggests that TPEN induces apoptosis independently of its metal chelator function. However, the source of TPEN generated O 2 •− is unknown. Whatever the mechanism of generation may be, we have shown that TPEN induces sustained generation of O 2 •− up to six hours of incubation and then slowly declined until complete incubation time (24 h   activation of NF-κB transcription factor in Jurkat cells showing fragmentation and condensation, typical indication of apoptotic morphology. Pharmacological inhibition of NF-κB with PDTC significantly inhibited TPEN-induced apoptosis. These observations comply with the notion that NF-κB is involved in apoptosis [46,47] in Jurkat cells exposed to TPEN. In line with this notion, NF-κB transcribes proapoptotic genes such as p53 [48]. TPEN induces p53 activation and translocation to nuclei. TPEN activates p53 not only in Jurkat cells (this work) but also in neuronal cells [28,33]. It is perhaps not surprising that activated p53 triggers apoptosis by altering mitochondria function [49], suppressing anti-oxidant genes [50] and regulating metabolic genes [51]. These data suggest that p53 acts both as crucial node downstream of diverse stress signals and as a stress sensor. However, p53-induced apoptosis can be stopped and reverse. Specific inhibitor PFT is able to protect and rescue TPEN-induced apoptosis in Jurkat cells. This observation further reinforces the notion that p53 is susceptible to regulation process, thereby potentially directed to specifically destroy malignant cells. P53 has therefore become an excellent candidate to therapeutic approaches against leukaemia cells [52][53][54]. In contrast to Ak and Levine [55], we conclude that both NF-κB and p53 have evolved to respond to OS and that they can function in the same cell at the same time. TPEN/H 2 O 2 stress provoked loss of ΔΨ m concomitantly with externalization of PtdSer in a time-dependent fashion, as typical marker of apoptosis. However, the flow cytometer values for loss ΔΨ m and PtdSer were significantly different. This observation suggests that either drop ΔΨ m occurs previous to externalization of PtdSer or Annexin V assay unsuccessfully label all apoptotic cells [56]. In accordance with others [25,28,[30][31][32][33], we found caspase-3 activation in treated cells. The participation of caspase-3 in TPEN toxicity was confirmed by using NSCI inhibitor, which not only protects but also rescues cells from TPEN toxicity. Caspase-3 activation therefore constitutes a critical protease and a marker of cell death induced by OS. We report for the first time that TPEN provokes AIF to translocate to the nucleus and induces chromatin condensation. In contrast to others [57], AIF is primarily involved in apoptosis in Jurkat cells exposed to TPEN, at least under the present experimental conditions. In support of this view, Stambolsky et al. [58] have shown that p53 regulates expression of AIF. We conclude that depending on the cell type and stress stimuli, AIF might be involved in apoptosis or necrosis [57,59]. It is concluded that TPEN induces apoptosis in Jurkat cells mainly mitochondrialmediated pathway [60] by two complimentary but independent cell death subroutes: AIF-and caspase-3 dependent mechanism. However, we found that the proportion of nuclei in stage II detected by fluorescence microscopy was higher than the proportion of nuclei in stage I. One possible explanation is that TPEN chelates zinc from caspase-3, increasing its catalytic activity [32] and thereby overpassing the mitochondrial caspase activation process. Furthermore, TPEN also destroy XIAP, a natural caspase-3 inhibitor [27]. It is therefore not surprising that zinc, being a potent inhibitor of caspase-3 [61], protects and rescues Jurkat cells from TPEN toxic effects. Therefore, AIF and caspase 3 should be used as regular markers of cell death. Additionally, TPEN induces apoptosis via activation of c-Jun and JNK kinase and their activation might be mediated by H 2 O 2 [62]. It has been reported that JNK phosphorylated p53 [63]. In view of the present data and those reported by Yin et al. [64] comply with the notion that NF-κB, JNK and p53 pathways are involved in OS in HepG2 cells and Jurkat cells (this work).