Acute lymphoblastic leukemia (aLL) is a devastating cancer of immature lymphocytes. It largely afflicts children, representing more than a quarter of all childhood cancers, and causing most of the fatalities from cancer in children (Hunger and Mullighan, 2015). L-asparaginase is a key component of aLL chemotherapy. Regimens that consist of L-asparaginase give rise to greater induction of remission compared to L-asparaginase-free regimens (Egler et al., 2016). L-asparaginase is thought to trigger asparagine insufficiency, causing protein synthesis inhibition and subsequent aLL cell death. However, treatment of L-asparaginase comes with the risk of resistance.

Using genome-wide RNA interference screening, we discovered huntingtin-associated protein 1 (HAP1) (Lee et al., 2019) as a novel biomarker for L-asparaginase resistance in aLL cells. Loss of HAP1 expression in aLL patient primary leukemic cells corresponds to L-asparaginase resistance, indicating that L-asparaginase induces aLL cell apoptosis (Kang et al., 2017; Lee et al., 2019) through a novel non-canonical pathway that involves HAP1. HAP1 binds to huntingtin (Htt) and the intracellular inositol 1,4,5- trisphosphate (IP3) receptor (IP3R) Ca²⁺ channel to form a functional HAP1-Htt-IP3R complex that regulates IP3-stimulated ER Ca²⁺ release. HAP1 loss inhibits HAP1-Htt-IP3R formation and thus L-asparaginase stimulation of ER Ca²⁺ release (Lee et al., 2019). Loss of HAP1 also reduces entry of external Ca²⁺, inhibiting an overwhelming increase in [Ca²⁺]_i, and downregulating the Ca²⁺-activated calpain 1, Bid, and caspase-3/12 apoptotic pathway, which result in L-asparaginase resistance (Lee et al., 2019). These findings indicate that L-asparaginase causes aLL cell apoptosis through perturbation of intracellular Ca²⁺ homeostasis and subsequent upregulation of the Ca²⁺-activated calpain 1, Bid, and caspase-3/12 apoptotic pathway. The ability of the Ca²⁺ chelator, BAPTA-AM, to almost completely reverse aLL cell apoptosis establishes an association between an increase in [Ca²⁺]_cyt and L-asparaginase-stimulated apoptosis (Lee et al., 2019). However, the cellular events that lead to a lethal increase in [Ca²⁺]_i following L-asparaginase-stimulated ER Ca²⁺ release are still unknown.

Mitochondria are cell organelles that regulate Ca²⁺ homeostasis and apoptosis. The outer mitochondrial membrane (OMM) is easily permeable to Ca²⁺ while the inner mitochondrial membrane (IMM) consists of the mitochondrial calcium uniporter (MCU) complex that mediates mitochondrial Ca²⁺ influx (Collins et al., 2001). As MCU has low affinity to Ca²⁺, increased Ca²⁺ concentration is required for MCU activity (Paupe and Prudent, 2018). Uptake of mitochondrial Ca²⁺ via MCU channels is made possible by the proximity between the mitochondria and the ER (Rizzuto et al., 2009; Grimm, 2012), the major intracellular Ca²⁺ store. A typical mechanism for ER-mitochondria communication is via the mitochondria-associated ER membrane (MAM) (Vance, 2014), the ER-mitochondria interface. MAMs are associated with several proteins such as IP3R Ca²⁺ channels (Patergnani et al., 2011) and voltage-dependent anion channels (VDACs) (Ma et al., 2017). These channels regulate ER Ca²⁺ transport to the mitochondria (Patergnani et al., 2011). Once ER Ca²⁺ is released through IP3R channels, mitochondria Ca²⁺ uptake occurs (Patergnani et al., 2011) via the OMM VDACs, and the IMM MCU channel (Rizzuto et al., 2009; Shoshan-Barmatz et al., 2017). However, overload of mitochondrial Ca²⁺ is related to not only increased or sustained formation of the mitochondrial permeability transition pore (mPTP) (Moore, 1971; Duchen, 2000; Finkel et al., 2015) but also the generation of mitochondrial ROS (Ermak and Davies, 2002; Gorlach et al., 2015), which also contributes to mPTP formation (Zorov et al., 2000; NavaneethaKrishnan et al., 2020) that allows ROS release into the cytoplasm (Zorov et al., 2014). The mPTP channel regulates the IMM permeabilization. Although transient mPTP opening serves as a mitochondrial Ca²⁺ efflux channel under normal conditions (Altschuld et al., 1992; Ichas et al., 1997), sustained mPTP formation triggers swelling of mitochondria and secretion of cytochrome C and other intermembrane space (IMS) proteins, causing caspase-regulated apoptosis (Lemasters et al., 2002; Kinnally et al., 2011).

In the current study, we utilized SEM patient aLL cells expressing or depleted of HAP1 by retroviral transfection, and demonstrate that L-asparaginase-induced aLL cell apoptosis triggered by a lethal rise in [Ca²⁺]_cyt is caused by mPTP formation that results from ER-mitochondria Ca²⁺ transfer and subsequent ROS production. Thus, our findings define the Ca²⁺-mediated mechanisms through which L-asparaginase perturbs intracellular Ca²⁺ homeostasis to cause apoptosis in aLL cells.

L-asparaginase-induced ER Ca²⁺ release that is mediated by IP3R causes mPTP formation. To investigate if L-asparaginase-induced IP3R-mediated ER Ca²⁺ release (Lee et al., 2019) causes mPTP formation, SEM cells (*) infected with retrovirus carrying an empty pRS vector (*+pRS) were loaded with calcein-AM and stimulated with L-asparaginase. Cells were then treated with CoCl₂ and analyzed by flow cytometry. Cell-permeable calcein-AM dye disperses and gets confined into subcellular organelles such as mitochondria (Petronilli et al., 1998). CoCl₂ removes calcein staining in all subcellular compartments except the mitochondria, which are surrounded by a CoCl₂-resistant inner mitochondrial membrane (IMM), when mPTP is closed (Petronilli et al., 1998). Thus, CoCl₂ treatment permits detection of status of mPTP formation (Petronilli et al., 1998). As shown in Figure 1A, L-asparaginase caused an obvious shift in calcein-stained population of *+pRS cells, indicating clear removal of calcein staining and, therefore, mPTP formation in these cells. To establish a link between L-asparaginase-induced IP3R-mediated ER Ca²⁺ release and mPTP formation, SEM cells (*) stably depleted of HAP1 by infection with retrovirus carrying pRS-shHAP1 (*+pRS-shHAP1) (Lee et al., 2019) were used. Lack of HAP1 (Figure 1B), a key component of the functional ER Ca²⁺channel, IP3R/HAP1/Htt ternary complex (Lee et al., 2019), inhibited L-asparaginase-induced ER Ca²⁺ release (Figure 1C). Treatment with TBHQ, an ER Ca²⁺ pump inhibitor, caused ER Ca²⁺ release in both *+pRS and *+pRS-shHAP1 cells, indicating viability of these cells during analysis. In *+pRS-shHAP1 cells where ER Ca²⁺ release was blocked due to HAP1 loss, L-asparaginase caused a modest shift in calcein-stained population (Figure 1A), indicating high and greater retention of calcein staining in these cells compared to *+pRS cells, and, therefore, closed mPTP. These findings indicate that L-asparaginase-induced IP3R-mediated ER Ca²⁺ release, which was observed in *+pRS cells, causes mPTP formation.

L-asparaginase-induced mPTP formation, which is dependent on IP3R-mediated ER Ca²⁺ release, results from Ca²⁺ entry into mitochondria. Since mPTP formation is associated with Ca²⁺ overload in mitochondria (Duchen, 2000; Contreras et al., 2010; Finkel et al., 2015; NavaneethaKrishnan et al., 2020) that could be mediated by the MCU located in the IMM (Moore, 1971), we tested the involvement of MCU in L-asparaginase-induced mPTP formation. To do so, *+pRS and ^*+pRS-shHAP1 cells pre-treated with Ruthenium Red (RuR), a potent MCU inhibitor (Moore, 1971), and stimulated with L-asparaginase were examined for mPTP formation as described above. As shown in Figures 2A, B, RuR increased calcein staining in *+pRS cells stimulated with L-asparaginase (left panel), indicating inhibition of L-asparaginase-induced mPTP formation. As expected, RuR had no effect on calcein fluorescence intensity in ^*+pRS-shHAP1 cells stimulated with L-asparaginase (right panel). Cyclosporine A (CsA), an mPTP inhibitor (Crompton et al., 1988), was used as positive control. These findings indicate that in aLL cells, L-asparaginase-induced mPTP formation, which depends on IP3R-mediated ER Ca²⁺ release, involves the MCU channel that is linked to mitochondrial Ca²⁺ uptake (Giorgi et al., 2018).

We then examined if L-asparaginase induces a rise in mitochondrial Ca²⁺ level ([Ca²⁺]_mt). To do so, *+pRS and *+pRS-shHAP1 cells loaded with the cell permeable mitochondrial Ca²⁺ dye, Rhod-2 AM³³, were treated with L-asparaginase, and analyzed for mitochondrial Ca²⁺ increase by spectrofluorometry. As shown in Figure 3, L-asparaginase, which causes ER Ca²⁺ release in *+pRS cells (Figure 1C), induced mitochondrial Ca²⁺ increase in these cells, but not in HAP1-depleted ^*+pRS-shHAP1 cells [where ER Ca²⁺ release is inhibited (Figure 1C)]. To further establish a link between L-asparaginase-induced IP3R-mediated ER Ca²⁺ release and increased [Ca²⁺]_mt, *+pRS cells were pre-treated with Xestospongin C (XeC), a potent inhibitor of IP3R (Gafni et al., 1997), prior to L-asparaginase treatment, and spectrofluorometric Ca²⁺ analysis was performed. XeC dramatically reduced mitochondrial Ca²⁺ increase in *+pRS cells, indicating that the rise in [Ca²⁺]_mt is associated with L-asparaginase-induced ER Ca²⁺ release and subsequent transfer to the mitochondria. Pre-treatment with RuR also inhibited L-asparaginase-induced ER-mitochondria Ca²⁺ transfer, further indicating the involvement of MCU in the process.

L-asparaginase-induced rise in [Ca²⁺]_mt evokes an increase in reactive oxygen species (ROS) in aLL cells. Since a rise in [Ca²⁺]_mt has been associated with the generation of mitochondrial ROS (Ermak and Davies, 2002; Gorlach et al., 2015), which also contributes to mPTP formation (Zorov et al., 2000; NavaneethaKrishnan et al., 2020) that allows ROS release into the cytoplasm (Zorov et al., 2014), we examined if L-asparaginase-induced rise in [Ca²⁺]_mt upregulates mitochondrial and cytosolic ROS levels in aLL cells. To determine mitochondrial superoxide anion levels, *+pRS and *+pRS-shHAP1 cells pre-treated with RuR, then stimulated with L-asparaginase were stained with MitoSOX and examined by microscopy. As shown in Figure 4, L-asparaginase induced a rise in mitochondrial ROS level in *+pRS cells, which was inhibited by pre-treatment with RuR and more so by HAP1 loss in *+pRS-shHAP1 cells. We then examined cytosolic ROS levels in *+pRS and *+pRS-shHAP1 cells pre-treated with RuR or CsA or Mito-Tempo, a mitochondrial ROS scavenger (NavaneethaKrishnan et al., 2018), then stimulated with L-asparaginase. Cells were stained with 2′,7′-dichlorofluorescin diacetate (DCFDA) and examined by microscopy. Figure 5 shows that L-asparaginase induced a rise in cytosolic ROS level in *+pRS cells, which was inhibited by pre-treatment with RuR, CsA or Mito-Tempo and more so by HAP1 loss in *+pRS-shHAP1 cells. These findings indicate that L-asparaginase-induced rise in [Ca²⁺]_mt is accompanied by mitochondrial and cytosolic ROS increases in aLL cells.

L-asparaginase-induced ER-mitochondria Ca²⁺ transfer and subsequent mPTP formation cause a rise in [Ca²⁺]_cyt. We then examined whether a rise in [Ca²⁺]_cyt observed in L-asparaginase-treated aLL cells (Lee et al., 2019) is due to ER-mitochondria Ca²⁺ transfer and subsequent mPTP formation. To do so, *+pRS cells loaded with Fluo-4 AM were pre-treated with XeC, RuR or CsA then treated with L-asparaginase and analyzed for Ca²⁺ transients by single-cell Ca²⁺ imaging. HAP1-depleted *+pRS-shHAP1 cells were again examined to test whether L-asparaginase-induced IP3R-mediated ER Ca²⁺ release is linked to a rise in [Ca²⁺]_cyt. As shown in Figure 6, L-asparaginase caused an increase in [Ca²⁺]_cyt in *+pRS cells, which was inhibited by pre-treatment with RuR or CsA and more so by XeC, and by HAP1 loss in *+pRS-shHAP1 cells. These findings indicate that IP3R-stimulated ER Ca²⁺ release and transfer to the mitochondria, and subsequent mPTP formation account for L-asparaginase-induced rise in [Ca²⁺]_cyt.

L-asparaginase-induced apoptosis is inhibited by blocking ER-mitochondria Ca²⁺ transfer, mPTP formation and/or mitochondrial ROS production. As indicated above, L-asparaginase causes aLL cell apoptosis by triggering IP3R-mediated ER Ca²⁺ release that results in a lethal rise in [Ca²⁺]_i and upregulation of the Ca²⁺-activated calpain-1-Bid-caspase-3/12 pathway (Lee et al., 2019). Treatment with the Ca²⁺ chelator, BAPTA-AM, in aLL cells reversed L-asparaginase-induced apoptotic cell death, indicating a link between [Ca²⁺]_cyt increase and apoptosis in aLL cells (Lee et al., 2019). Thus, we sought to determine whether L-asparaginase-induced ER-mitochondria Ca²⁺ transfer and subsequent ROS production, which cause mPTP opening that leads to increased [Ca²⁺]_cyt, are linked to L-asparaginase-induced apoptosis in aLL cells. To do so, *+pRS cells pre-treated with RuR, CsA or Mito-Tempo then stimulated with L-asparaginase were stained with Hoechst 34580 and FITC-Annexin V. FITC-positive apoptotic cells were counted 12 h post-treatment and the percentage of apoptotic cells was determined. As shown in Figure 7, L-asparaginase-induced aLL cell apoptosis was inhibited by RuR, CsA or Mito-Tempo in *+pRS cells (left panel) but not in *+pRS-shHAP1 cells (right panel), which showed no ER Ca²⁺ release upon stimulation with L-asparaginase (Figure 1C). Altogether, our findings indicate that L-asparaginase-induced aLL cell apoptosis caused by an IP3R-mediated ER Ca²⁺ release and subsequent lethal rise in [Ca²⁺]_cyt (Lee et al., 2019) results from ER-mitochondria Ca²⁺ transfer and ROS production, which lead to mPTP formation.

Mitochondria regulate a number of cellular processes, including Ca²⁺ homeostasis and apoptosis. They communicate dynamically with the ER and store part of the released ER Ca²⁺. In this study, we demonstrate that L-asparaginase-induced IP3R-mediated ER Ca²⁺ release in aLL cells causes mPTP formation, which is inhibited in cells lacking HAP1 or upon inhibition of IP3R. These findings establish a link between L-asparaginase-induced IP3R-mediated ER Ca²⁺ release and mPTP formation. RuR inhibition of L-asparaginase-induced mPTP formation indicates the involvement of the MCU channel that is important for mitochondrial Ca²⁺ uptake (Giorgi et al., 2018). The fact that L-asparaginase also evokes [Ca²⁺]_mt increase in aLL cells, but not in those depleted of HAP1 or upon inhibition of IP3R suggests a link between L-asparaginase-induced Ca²⁺ transfer from the ER to the mitochondria and the rise in [Ca²⁺]_mt.

Our finding that L-asparaginase-induced rise in [Ca²⁺]_mt is accompanied by mitochondrial and cytosolic ROS increase in aLL cells is consistent with the notion that a rise in [Ca²⁺]_mt contributes to mitochondrial ROS production (Ermak and Davies, 2002; Gorlach et al., 2015), which stimulates mPTP formation (Zorov et al., 2000; NavaneethaKrishnan et al., 2020) that facilitates the release of ROS into the cytoplasm (Zorov et al., 2014). Although we observed significant RuR inhibition of L-asparaginase-induced rise in both [Ca²⁺]_mt and mitochondrial ROS level in *+pRS cells, the degree of inhibition of mitochondrial ROS level is less than that of [Ca²⁺]_mt. This difference may arise from the different time of measurement: Ca²⁺ response occurs within seconds and thus was measured immediately; on the other hand, ROS response is slower and was measured 12 h following L-asparaginase treatment. In addition, since ROS is produced in mitochondria and leaks into the cytoplasm through mPTP, there will be dynamic changes in mitochondrial levels of ROS which eventually accumulates in the cytoplasm. This explains the similar degree of RuR inhibition of [Ca²⁺]_mt and cytoplasmic ROS levels. Thus, differences in the extent of RuR inhibition in [Ca²⁺]_mt and mitochondrial ROS can be attributed to differences in method and time of measurement. Non-etheless, it is clear that MCU-mediated inhibition of [Ca²⁺]_mt increase also causes inhibition of ROS increase in both mitochondria and cytoplasm.

As for our view that the rise in [Ca²⁺]_cyt results from L-asparaginase-induced IP3R-mediated ER Ca²⁺ release and transfer to the mitochondria and subsequent mPTP formation, this is substantiated by the observed inhibition of the process in cells lacking HAP1 or when ER Ca²⁺ release, MCU, or mPTP is inhibited. Our finding that cells with inhibited MCU show greater [Ca²⁺]_cyt compared to cells with inhibited IP3R or depleted of HAP1 suggests that MCU inhibition causes released ER Ca²⁺ to bypass the mitochondria and go directly into the cytosol.

Overall, our findings align with previous studies showing that loss of cyclin-dependent kinase 5 (Cdk5) in breast cancer cells or knocking out Cdk5 in primary mouse embryonic fibroblasts (MEFs) is associated with mPTP formation, ROS increase [Ca²⁺]_mt and [Ca²⁺]_cyt increase, and caspase-mediated apoptosis (NavaneethaKrishnan et al., 2018; NavaneethaKrishnan et al., 2020). Our notion that L-asparaginase-induced apoptosis in aLL cells involves ER-mitochondria Ca²⁺ transfer, ROS production and mPTP formation is validated by inhibition of apoptosis upon inhibition of MCU channel, ROS production and/or mPTP formation.

Since stimulation of µ-opioid receptors (µ-ORs) was shown to trigger G_βγ-mediated rise in [Ca²⁺]_cyt through phospholipase C (PLC) (Yoon et al., 1999; Charles et al., 2003; Celik et al., 2016; Machelska and Celik, 2018; Lee et al., 2021), we propose a model (Figure 8) whereby L-asparaginase causes aLL cell apoptosis via our previously identified Ca²⁺-mediated calpain-1-Bid-caspase-3/12 apoptotic pathway (in purple) (Lee et al., 2019). In this model, we show that L-asparaginase stimulates GPCR (e.g., PAR2 (Peng et al., 2016)) in aLL cells, causing G_βγ-stimulation of PLC, which causes a rise in [Ca²⁺]_cyt through IP3R-mediated release of ER Ca²⁺ and mitochondrial Ca²⁺ uptake that leads to ROS production, which together induce mPTP formation.

In conclusion, our findings indicate that L-asparaginase-induced aLL cell apoptosis requires ER-mitochondria Ca²⁺ transfer, ROS production and mPTP formation. Thus, results from our studies not only fill in the gaps in our understanding of the Ca²⁺-mediated mechanisms by which L-asparaginase induces aLL cell apoptosis, but also offer a fresh perspective on targeting ER Ca²⁺ release, ER-mitochondria Ca²⁺ transport, ROS and/or mPTP in leukemic cells for aLL therapy.