Loss of the redox mitochondrial protein mitoNEET leads to mitochondrial dysfunction in B-cell acute lymphoblastic leukemia

Abstract

B-cell acute lymphoblastic leukemia (ALL) affects both pediatric and adult patients. Chemotherapy resistant tumor cells that contribute to minimal residual disease (MRD) underlie relapse and poor clinical outcomes in a sub-set of patients. Targeting mitochondrial oxidative phosphorylation (OXPHOS) in the treatment of refractory leukemic cells is a potential novel approach to sensitizing tumor cells to existing standard of care therapeutic agents. In the current study, we have expanded our previous investigation of the mitoNEET ligand NL-1 in the treatment of ALL to interrogate the functional role of the mitochondrial outer membrane protein mitoNEET in B-cell ALL. Knockout (KO) of mitoNEET (gene: CISD1) in REH leukemic cells led to changes in mitochondrial ultra-structure and function. REH cells have significantly reduced OXPHOS capacity in the KO cells coincident with reduction in electron flow and increased reactive oxygen species. In addition, we found a decrease in lipid content in KO cells, as compared to the vector control cells was observed. Lastly, the KO of mitoNEET was associated with decreased proliferation as compared to control cells when exposed to the standard of care agent cytarabine (Ara-C). Taken together, these observations suggest that mitoNEET is essential for optimal function of mitochondria in B-cell ALL and may represent a novel anti-leukemic drug target for treatment of minimal residual disease.

Graphical Abstract

1. Introduction

MitoNEET (gene CISD1) is an iron-sulfur containing protein located on the outer mitochondrial membrane [1]. This evolutionally conserved zinc finger protein contains a CDGSH-iron-sulfur domain [1–3] where the iron sulfur [2Fe-2S] clusters are redox active and thought to contribute to the redox and pH-sensor regulation of OXPHOS bioenergetics in mitochondria [1, 4–6]. Originally discovered as an off-target effector for the anti-diabetic glitazone drug pioglitazone, mitoNEET was found to have therapeutic benefits on β-oxidation and lipid metabolism unrelated to its peroxisome proliferator-activated receptor (PPAR)-γ activation [7–9]. MitoNEET has been identified in bioinformatic and functional studies to be associated with several cancers, including breast and lung cancer [10–12]. In breast cancer for instance, mitoNEET knockout led to a dramatic reduction in solid tumor size, and studies suggested that ligands targeting mitoNEET may modulate cancer proliferation and chemo-resistance [13, 14].

Cancer cells exhibit differential metabolism compared to normal cells [15–17], and while it is known that cancer cells are mainly glycolytic in nature (Warburg effect), alterations in mitochondrial oxidative phosphorylation (OXPHOS) can contribute to changes in chemo-resistance through mechanisms that remain to be fully elucidated [18]. Targeting mitochondrial OXPHOS in cancer cells has been suggested as a novel approach to eradicate minimal residual disease (MRD) in leukemia patients [19, 20]. MRD persists in some B-cell acute lymphoblastic leukemia (ALL) patients following completion of chemotherapy treatment, with the presence of surviving cells being a negative prognostic indicator for patients. The presence of MRD posttreatment has been attributed to leukemic cell populations residing in the bone marrow microenvironment (BMM), due to changes in cellular metabolism [21, 22]. This unique anatomical microenvironment is considered a “site of sanctuary” for leukemic cells and contributes to disease relapse and chemo-resistance through multiple metabolic pathways [21, 23].

Our previous studies of a first-in-class mitoNEET ligand, NL-1, showed activity in B-cell ALL, which prompted interest in evaluating mitochondrial bioenergetic function in the setting of mitoNEET loss, to evaluate the potential of leveraging this pathway for novel therapeutic benefit in B-cell ALL [24].

2. Material and Methods

2.1. Cell culture and materials

Human derived REH cells (ATCC #CRL-8286) were purchased and maintained in RPMI 1640 supplemented with 10% FBS and 1x streptomycin/penicillin antibiotics. MitoNEET (gene: CISD1) knockout (KO) cells were generated by Synthego (Menlo Park, CA) using the CRISPR/Cas9-based gene edition system in REH cells. The guide sequence was GAUCUCCACAUAGGGGCCAG. The PCR primers used were TTCTAGGGAGCAGGTCCTGACT (forward) and TCCTTTCCTACAAGCTCGTGGG (reverse). The control cells used throughout the study were the empty vector transduced control REH cells (vector control; VC).

2.2. GFP labeled cells

REH cells (5,000 cells/100 μl) were placed in fresh medium containing 5 μg/ml Polybrene in a 96 well plate. Firefly Luciferase-F2A-GFP Lentivirus (Fluc-F2A-GFP; Biosettia, San Diego, CA) was added (5 μl/well). The plate was spun at 1,000 x g for 60 min at room temperature then placed in 37° C, 5% CO₂ incubator for 24 hrs. The next day, the media was replaced with fresh media to remove lentivirus and Polybrene. After 12 hrs, 500 μg/ml hygromycin was added for 72 hrs to select for transduced cells.

2.3. Cell proliferation studies

WST-8 assay proliferation assay. Viable cells were quantitated using the Cell Counting Kit-8 (Dojindo Molecular Technology Inc., cat#CK04) according to manufacturer’s instructions at indicated time points. Briefly, 10 μl of the assay reagent was added to each well of the 96-well plate and incubated for 2 hrs at 37 °C, after which the plates were read on a BioTek Cynergy 5 plate reader at 450 nm absorbance. MitoNEET KO cells were compared to VC cells for analysis. In addition, GFP-labeled REH proliferation studies were completed by plating fluorescent REH VC and KO cells at 50,000/well in 100 μl/well followed by reading the plate at the indicated time points using a Molecular Devices SpectraMax ID5 plate reader.

2.4. Ara-C studies

Cells were plated in 96-well plates at a density of 50,000 cells/well in 100 μl. Cells were treated with Ara-C at indicated concentrations, and cell proliferation was measured after 48 hrs of incubation with the compounds using a Cell Counting Kit-8 (Dojindo Molecular Technology Inc. Cat#CK04), which was used according to manufacturer’s instructions. Briefly, 10 μl of the assay reagent was added to each well and incubated for 2 hrs at 37 °C, after which the plates were read on a BioTek Cynergy 5 plate reader at 450 nm absorbance. Untreated cells were used as controls.

2.5. ATP studies

Cells were plated in 96-well plates at a density of 50,000 cells/well in 100 μl. ATP was quantitated using the CellTiter-Glo 2.0 Assay (Promega Cat#G9241) according to manufacturer’s instructions and read using a Molecular Devices SpectraMax ID5 plate reader.

2.6. Western blot analysis

MitoNEET VC and KO ALL cells were cultured in suspension prior to lysis in RIPA buffer, and protein concentrations were determined using the bicinchoninic acid (BCA) protein assay kit (Cat#23225 Pierce-ThermoFisher Scientific, Waltham, MA). Equal quantities of proteins were resolved on 4-20% Mini-Protean TGX Precast Protein Gels (BioRad Cat#45610994) and transferred to nitrocellulose membranes. Membranes were blocked in TBS/0.1% Tween-20 with 5% nonfat dry milk for 1 hr at RT. Blots were probed with the indicated primary antibodies overnight then rinsed with TBS/0.1% Tween-20. Blots were incubated with horseradish peroxidase-conjugated secondary antibodies for 1 hr at RT, then rinsed 3X with TBS/0.1% Tween-20. Signal was visualized using Immobilon Western Chemiluminescent HRP Substrate (Millipore Cat#WBKLS0500) on the Amersham Imager AI680. The following antibodies CISD1/MitoNEET (#83775), SOD2 (#13141), GPX1 (#3286), GPX4 (#52455) and β-actin (#8457) were purchased from Cell Signaling Technologies (Danvers, MA). Western blots are representative of at least three independent experiments.

2.7. Microscopy

For the MitoTracker and mitoNEET co-localization study, cells were stained with 100 nM MitoTracker Deep Red cat #M22426 for 30 min at 37 ° C. Cells were washed with 1X PBS then fixed in 4% PFA for 15 min at 37 ° C. Cells were washed with 1X PBS then incubated with 0.4% Triton X-100 in PBS for 10 min at RT. Cells were washed 1X in PBS before adding mitoNEET antibody (Cell signaling Technologies cat #83775), 1 μl diluted in 200 μl of PBS + 5% BSA, for 1hr at RT. The cells were washed with 1X PBS before adding donkey anti-rabbit AF555 diluted 1:400 for 1 hr at RT. Cells were then washed 2 times in PBS before slides were made and Prolong with DAPI was used prior to adding a coverslip. The SIM (Structured Illumination Microscopy) images were acquired using a Nikon A1R w/ N-SIM E microscope, using the 100X/1.49 SR objective. For the TOM20 and mitoNEET co-localization study, cells were counted, washed with 1X PBS, then fixed in 4% PFA for 15 min at 37 ° C. Cells were washed with 1X PBS then incubated with 0.4% Triton X-100 in PBS for 10 min at RT. Cells were washed 3X with 1X PBS then incubated for 1 hr at RT with 5% goat serum in 1X PBS. Blocking buffer was removed then 1:200 dilution of TOM20 Coralite 488-conjugated polyclonal antibody (Proteintech cat#CL488-11802) added for 1.5 hr at RT. They were washed 3X in 1XPBS before adding mitoNEET antibody (Cell Signaling Technologies cat #83775), 1 μl diluted in 200 μl of PBS + 5% BSA, for 1hr at RT. The cells were washed with 1X PBS before adding donkey anti-rabbit AF555 diluted 1:400 for 1 hr at RT. They were then washed 2X in 1X PBS before slides were made and Prolong with DAPI was used prior to adding a coverslip. The images were taken using the 100X/1.49 SR objective on the Nikon A1R confocal. For the Nile Red assay, following isolation of REH cells, 4 million cells were re-suspended in PBS containing 2 μM of Nile Red (Invitrogen, Carlsbad, CA) for 10 min. Following incubation, the cells were rinsed with PBS, re-suspended in PBS, and cytospun on slides. The Nile Red images were taken on the Nikon A1R confocal microscope using the 60X/1.4 objective.

2.8. Metabolite characterization

To determine levels of glutathione (GSH) and GSSH, a luciferase-based GSH/GSSG assay from Promega (cat# V6611) was used, according to the manufacturer’s protocol. NADP/NADPH levels were measured using a luciferase-based kit from Promega (cat# G9081) according to the manufacturer’s protocol. Cells were seeded at 50,000/well in 100 μl/well in 96-well white plates with an opaque bottom. Luminescent readings were completed using a Cytation 5 multi-modal plate reader (BioTek).

2.9. Electron microscopy

A cell suspension was fixed in glutaraldehyde and paraformaldehyde solution for 1 hr before being pelleted by centrifugation. Pellets were enrobed in agarose and then stained with osmium tetroxide (OsO₄) solution. The agarose pieces containing cells were then gradually dehydrated, infiltrated, and embedded in EPON resin. Ultra-thin sections (70-90 nm) were collected onto copper grids and further stained with uranyl acetate and lead citrate. Sections were imaged under TEM at 80KV to visualize mitochondrial ultrastructure.

2.10. Cellular respiration

A MitoStress Test assay, utilizing the Seahorse XFe96 Bioanalyzer (Agilent Technologies, Santa Clara, CA) was used to assess the impact of mitoNEET on cellular respiration. One hour prior to the start of the assay, MitoNEET KO and VC REH cells, suspended in Mito-Stress assay media containing glucose, L-glutamine and sodium pyruvate, were seeded into Seahorse 96-well XF cell culture microplates coated with Cell-Tak Cell and Tissue Adhesive (Corning Inc., Glendale, AZ) at a density of 100,000 cells/well. Cell-Tak solution contained a ratio of 289 μl of 0.1 M sodium bicarbonate, 1 μl of 5N NaOH, and 10 μl of Cell-Tak, with 10 μl added to wells, then discarded after 20 mins by 2 washes of 100 μl of water. For the Glycolysis Stress Test, media containing 49.5 mL of sterile XF DMEM Medium, pH 7.4 with 500 μl of Seahorse XF 200 mM glutamine was added to the cells. The 96-well plate was centrifuged at 200 x g for one minute and placed in a non-CO₂ incubator for 1 hr prior to beginning the experiment. The Seahorse XFe96 measured oxygen consumption rate (OCR) under specific conditions using sequential addition of oligomycin, carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP), antimycin-A and rotenone for the MitoStress Test. For the Glycolysis Stress test, glucose, oligomycin and 2-deoxygclucose were added as per kit instructions.

2.11. Electron flow assay

An Electron Flow Assay utilizing the Seahorse XFe96 Bioanalyzer (Agilent Technologies, Santa Clara, CA) was used to assess the impact of mitoNEET on activities of ETC complexes in permeabilized cells. REH VC and mitoNEET KO cells were seeded into Seahorse 96-well XF cell culture microplates coated with Cell-Tak Cell and Tissue Adhesive (Corning Inc., Glendale, AZ) at a density of 300,000 cells/well 1 hr before the assay was performed. The plate was centrifuged at 200 x g for 60 sec and placed in a non-CO₂ incubator for 1 hr prior to the start of the experiment. The XFe96 Bioanalyzer was calibrated with the following injections added to their respective injection ports: Port A: rotenone (2 μM); Port B: succinate (10 mM); Port C: antimycin-A (4 μM) and Port D: ascorbate (10 mM) + cytochrome c electron donor N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD; 100 μM). Pyruvate (10 mM), malate (1 mM), FCCP (4 μM) and Seahorse XF Plasma Membrane Permeabilizer (2 nM) were added to the XFe96 cell plate immediately before beginning the assay. Two mix-wait-measure cycles consisting of 30 sec mix, 30 sec wait, and 2 min measure were performed before the first injection and following each injection. ETC complex I activity was calculated as the complex I substrate (pyruvate + malate)-driven oxygen consumption sensitive to rotenone inhibition. ETC complex II/III activity calculates as the complex II/III substrate (succinate)-driven oxygen consumption sensitive to antimycin-A inhibition.

2.12. Superoxide assay with dihydroethidium bromide (DHE)

Cells were plated in 96-well V-bottom plate at a density of 100,000 cells/well in 100 μl. DHE was quantitated using a superoxide ROS Detection Cell-Based Assay Kit (Cayman Chemical Cat#601290) according to manufacturer’s instructions and read using a Molecular Devices SpectraMax ID5 plate reader.

3. Results and Discussion

MitoNEET is a redox active [2Fe-2S] cluster-containing protein which has been shown to regulate the oxidative capacity of mitochondria and cellular iron homeostasis [25]. Bioinformatic analyses and studies of cancers including breast, lung, and prostate cancer have reported that mitoNEET over-expression may contribute to tumor growth, and the knockdown or knockout of this protein reduces cell proliferation and/or tumor size, demonstrating a potential role for mitoNEET in cancer growth [10, 12, 14, 26]. MitoNEET-overexpression has also been shown to attenuate the effects of oxidative stress in cardiomyocytes and other cell types [27–30]. We determined that mitoNEET is expressed in primary patient samples isolated during relapse of B-ALL (Supplemental Fig. 1). We also recently showed that the mitoNEET ligand, NL-1, exhibited anti-proliferative activity in human REH cells, prompting our investigation into the role of mitoNEET in mitochondrial function in REH leukemic cells [24]. The need for novel approaches to treating MRD in ALL and potential role of mitoNEET in tumor progression, led us to postulate that mitoNEET ligands may provide benefit in ALL treatment by influencing mitochondrial metabolism.

To evaluate the role of mitoNEET in leukemic cells, we generated a mitoNEET knockout (KO) cell-line in REH B-cell ALL cells using CRISPR-Cas 9 methodology (Fig. 1A). Since mitoNEET has previously been shown to localize on the outer mitochondrial membrane of mitochondria [5, 6, 31], we confirmed through co-staining with MitoTracker that mitoNEET is localized with the mitochondria in the VC REH cells, and mitoNEET staining is lost in the mitoNEET KO cells (Fig. 1B) consistent with previously published studies [5, 6]. As an additional orthogonal evaluation of mitoNEET localization, mitoNEET was shown to localize with the translocase of the outer membrane protein (TOM20), a well-known mitochondrial marker (Supplemental Fig. 2). To visualize the mitochondrial ultra-structure in more detail, TEM imaging was employed and demonstrated that the mitochondrial ultrastructure was altered in mitoNEET KO cells (Fig. 1C). Ultra-structural alterations in mitochondria where mitoNEET was knocked-out, have also been noted by other groups, with noticeable decrease in the complexity of the cristae structure located in mitoNEET KO cellular mitochondria [1, 32]. Since mitochondrial function is closely tied to the size and shape of the mitochondria [33], we expected the mitoNEET KO mitochondria to have diminished functional capacity, specifically the bioenergetic OXPHOS capacity of the leukemia cells [32].

Fig. 1.

A) The mitochondrial protein mitoNEET was knocked out (KO) of REH cells using CRISPR/Cas9 for comparison to vector control cells (VC) and the protein expression determined with Western Blot. Beta-actin was used as the loading control (N = 3). B) Cellular localization of mitoNEET in REH cells. MitoNEET (red) is located on mitochondria, which are stained with MitoTracker Deep Red (green). Co-localization is observed from the yellow/orange coloring in the overlay in the VC cells. MitoNEET KO cells do not show localized staining in the mitochondria. Scale bar = 2 μm. C) TEM micrographs of REH cells. The KO cells are characterized by structural changes in the size and shape of the cristae in mitochondria, which are indicated by arrows, compared to the VC. Scale bar = 500 nm.

Based on previous studies showing mitoNEET regulates mitochondrial bioenergetic function [1, 29, 34], we utilized the Seahorse Bioanalyzer to characterize mitochondrial function including OXPHOS (measured by the oxygen consumption rate; OCR) and glycolysis (measured by the extracellular acidification rate; ECAR). MitoNEET knockout in REH cells was associated with reduction in overall mitochondrial function, including diminished non-mitochondrial oxygen consumption, basal respiration, maximal respiration, proton leak, ATP production, and spare respiratory capacity (Fig. 2A–F). Fig. 2G and 2H show an overall reduction in OCR and a nonsignificant decrease in ECAR. When the cells were analyzed using the Glycolysis Stress test, no significant differences were found between the KO and VC cells (Fig. 2I), even though the total ATP content in the KO cells was significantly reduced (Fig. 2J) [19, 35, 36]. These findings are consistent with earlier work focused on cardiac and neuronal cells showing the loss of mitoNEET in mitochondria greatly affected mitochondrial bioenergetic (OXPHOS) function, although loss of mitoNEET did not result in a complete shutdown of the electron transport chain (ETC) function [30, 34]. As a final evaluation of mitochondrial function, we assessed electron flow rates through the ETC (Fig. 3A and 3B). In comparison to the VC cells, there was a significant decrease in electron flow rates in the KO cells. Complex I (Fig. 3C), as well as the Complex II/III activity (Fig. 3D) was reduced, providing additional evidence that a loss of mitoNEET in leukemia cells impairs the OXPHOS capacity function of mitochondria.

Fig. 2.

Mitochondrial function in REH leukemia cells is disrupted. A mitochondrial stress test kit with the Seahorse bioanalyzer shows reduction in several key aspects of mitochondrial function including: A) non-mitochondrial oxygen consumption, B) basal respiration, C) maximal respiration, D) proton leak, E) ATP production, F) spare respiratory capacity; G) oxygen consumption (OCR) graphs of mitochondrial function. H) The extrasellar acidification rate (ECAR) graph, and I) Glycolysis stress test (no significant difference). J) ATP content (μM) is reduced in the mitoNEET KO cells. Data are shown average ± SD, where N = 8. *P<0.05.

Fig. 3.

Mitochondrial electron transport chain (ETC) function is diminished in mitoNEET KO REH cells. A) OCR graph showing substrate and inhibitor addition, B) averaged OCR response, C) Complex I Activity from the ETC data and D) Complex II/III activity. Data are shown as the average ± SD, where N = 8. *P<0.05. Abbreviations: rotenone (rot), succinate (sue), antimycin A (AA), ascorbate (asc), and N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD).

The mitochondrial OXPHOS and electron flow have previously been shown to be influenced by a loss of mitoNEET [37], with overexpression of mitoNEET leading to increased mitochondrial OXPHOS [38]. These changes in mitochondrial function in REH cells prompted us to evaluate other parameters of mitochondrial function including reactive oxygen species (ROS) production and mitochondrial membrane potential. In REH cells, the KO of mitoNEET led to increased levels of hydrogen-peroxide ROS production measured by the Amplex Red assay kit (Fig. 4A), and a decrease in the mitochondrial membrane potential (Fig. 4B), with the latter being a contributor of mitochondrially-derived ROS [39]. We found that the basal level of superoxide was significantly increased in the KO cells as measured by the superoxide indicator dye DHE, and as expected, antimycin A, an inhibitor of the ETC complex III, induced a significant increase in superoxide load in the KO cells as compared to VC cells (Fig. 4C). The antioxidant N-acetyl-cysteine was able to significantly reduce the superoxide levels in both the KO and VC cells. Combined with the results from the Mito-Stress and electron flow assays, these findings support a potential role for mitoNEET in the regulation of mitochondrial production of ATP under normal physiological parameters in concert with low production of noxious radicals. The absence of mitoNEET in leukemia cells leads to mitochondrial dysfunction and identifies a potential vulnerability that could be manipulated during treatment.

Fig. 4.

MitoNEET KO leads to mitochondrial dysfunction in REH cells. A) hydrogen peroxide ROS levels are increased in KO cells compared to VC; B) mitochondrial membrane potential is decreased in KO cells compared to VC; C) DHE detection of superoxide indicates increased levels in the KO cells compared to VC. The positive control antimycin A increased superoxide in both the VC and KO cells. The antioxidant N-acetyl cysteine (NAC) decreased both endogenous as well as antimycin A-induced superoxide production. Data are shown as the average ± SD, where N = 3-4. *P<0.05. D) Thiol content in REH cells with levels of reduced (GSH) and E) oxidized (GSSH) glutathione were measured in the KO cells compared to the VC. The loss of mitoNEET was correlated with a decrease in thiol content. F) Western Blot analysis of glutathione peroxidase (GPX) 1 and 4 as well as superoxide dismutase 2 (SOD2) showed a decrease in protein content in the mitoNEET KO cells compared to the VC. Equal protein loading per lane verified with β-actin staining. Data are shown average ± SD, where N = 8. *P<0.05. G) Metabolites NADP and NADPH in REH cells show significant increase in mitoNEET KO cells compared to the controls (VC). RLU average ± SD, where N = 8. *P<0.05.

The [2Fe-2S] clusters of mitoNEET are redox active, similar to other Fe-S containing proteins [40, 41], and supports the maintenance of free thiol content of cells including the antioxidant glutathione [42]. Electron paramagnetic resonance (EPR) studies further show that mitoNEET interacts with cytosolic metabolites such as glutathione, and NADPH/NADH [43]. Since these metabolites are essential for supporting the redox reactions in cellular processes, we evaluated their levels in the mitoNEET KO cells. Fig. 4D shows that the level of glutathione in the KO cells was dramatically reduced when compared to the VC cells, while no difference was observed in the levels of the oxidized GSSH (Fig. 4E). We also found that the protein levels of glutathione peroxidase (GPX) 1 and GPX4 were reduced in the KO cells, as well as the manganese-dependent superoxide dismutase (Mn-SOD; SOD2) (Fig. 4F). GPX1 is mainly located in the cytosol of most cell types, while GPX4 (also known as phospholipid hydroperoxide) is located in the mitochondrial and cytosolic spaces, and plays a role in lipid metabolism [39, 44]. The observed reduction of protein levels might be due, in part, to the higher load of ROS produced by the cells and decreased interaction with mitoNEET [42], and may also play a role in ferroptosis and iron dysregulation pathways [11, 45, 46]. Additionally, previous studies have indicated that some leukemias have increased levels of NADPH, which plays an important role in redox mechanisms [47, 48], and may explain the resulting increases in ROS detected. NADH+ and NADPH levels are dramatically increased in the KO cells (Fig. 4G), suggesting that reduced mitoNEET expression impacts the metabolism of NADH/NADPH and the metabolic signaling of these factors.

It was previously suggested that mitoNEET may play a role in lipid metabolism since patients taking pioglitazone had decreased free fatty acid plasma levels, while overexpression of mitoNEET in an obese mouse model led to significant improvements in lipid metabolism [7, 29, 30, 49]. Furthermore, the glitazone derivative PNU-91325 increased fatty acid synthesis in liver HepG2 cells [9]. Recently, it has been suggested that metabolic dysregulation may contribute to MRD, and taken together with the mitoNEET-associated effect on lipids alterations, we assessed the level of lipids in the leukemia KO and VC cells using Nile Red lipid staining [50]. As demonstrated in Fig. 5A, loss of mitoNEET led to a significant decrease in lipid levels in the cells. In order to explore this relationship between mitoNEET and fatty acids [9, 29], we evaluated the levels of free fatty acids via a triglyceride assay, but did not find any significant differences between the KO and the VC cells (Fig. 5B). Lastly, evaluation of the fatty acid metabolism utilizing the Seahorse BioFlux Analyzer evaluated the effect of mitoNEET KO on β-oxidation in the REH cells (Fig. 5C). As part of the assay, etomoxir C, an irreversible inhibitor (due to the covalent binding epoxide moiety) to the carnitine palmitoyltransferase-1, was used, as it prevents the formation of the acyl carnitines which is necessary for the production of ATP from fatty acid oxidation [51]. We found that the KO cells had lower levels of fatty acid oxidation when treated with etomoxir C, but this decrease was not statistically significant. Future studies will evaluate the β-oxidation pathways in ALL, especially those involved with Co-A derivatives, as fatty acids may play a role in mitochondrial autophagy [9, 29, 52].

Fig. 5.

mitoNEET KO leads to a decrease in cellular lipid levels. A) Nile Red staining shows lipid levels and DAPI nuclear staining, scale bar = 20 μm; B) free fatty acid (FFA) levels are unchanged with mitoNEET KO, C) β-oxidation assay with the Seahorse BioFlux Analyzer show that OCR is reduced in mitoNEET KO cells, average ± SD, where N = 8.

Mitochondrial metabolism in cancer cells has been suggested as a potential novel target for anticancer drug development [38]. MitoNEET has been suggested to play a role in cancer cell proliferation, where previous studies in breast cancer showed that elevated levels of mitoNEET led to increased cell proliferation [14, 53]. We evaluated the effect mitoNEET KO on REH cell proliferation and found that KO cells showed reduced proliferation (Fig. 6 and Supplemental Fig. 3), and a similar effect on cell proliferation was seen in our previously published studies showing that NL-1, a mitoNEET ligand, reduced cell proliferation in a B-cell ALL cancer model [24]. We next evaluated the effect of mitoNEET KO, in combination with the anti-cancer drug cytarabine (Ara-C), a common treatment of leukemia. Ara-C is an antimetabolic agent which interferes with the normal synthesis of DNA [54]. We found that treatment with Ara-C led to a significant decrease in cell proliferation in KO cells compared to the VC cells (Fig. 7), suggesting that the increased ROS-based stress on the cell with mitoNEET KO led to increase susceptibly of the cells to Ara-C. Since mitochondrial metabolism is needed for tumor growth to support proliferation [38], several groups have suggested that targeting mitochondria might be a useful strategy in cancer therapy. However, this strategy has been explored predominantly in solid tumors with insight into hematological cancers remaining less explored [55, 56]. By exploiting this metabolic vulnerability in leukemias, future combination therapies, including therapies targeting mitochondria (e.g. metformin [38]) or specifically mitoNEET, might provide additional benefit to patients.

Fig. 6.

MitoNEET KO results in decreased proliferation of REH cells compared to control cells (VC). Data are shown average ± SD, where N = 4. *P<0.05.

Fig. 7.

MitoNEET KO in REH cells increases sensitivity to the ALL anti-cancer drug Ara-C. Proliferation was significantly decreased in the KO cells compared to the VC. Data are shown average ± SD, where N = 4. *P<0.05.

In conclusion, we have characterized the effects of mitoNEET KO on mitochondrial bioenergetic function in human REH leukemic cells. Loss of mitoNEET dramatically affected mitochondrial bioenergetic function, decreased cellular OXPHOS, including dysregulated ROS generation, and decreased tumor cell proliferation. Based on these findings, targeting mitochondrial metabolism in B-ALL cancer cells through interruption of mitoNEET function provides a possible novel approach to of the treatment refractory MRD. Future studies will interrogate the potential of targeting mitochondrial function to sensitize treatment-refractory ALL to currently available chemotherapeutic agents.

Supplementary Material

1

Supplemental Fig. 1. MitoNEET protein levels in primary cancer specimens from two B-cell ALL patients. Equal protein loading per lane verified with β-actin staining.

Supplemental Fig. 2. Co-localization of mitoNEET (red) with TOM20 (green), a mitochondrial marker, with yellow areas showing overlap in REH cells. DAPI was used for nuclear staining. Scale bar = 2 μm.

Supplemental Fig. 3. Cell proliferation in the REH KO and VC cells was determined by measuring the fluorescence in GFP-expressing cells. *P<0.05, where N = 3.

Highlights.

• MitoNEET is a mitochondrial redox-active [2Fe-2S] cluster protein

• Knockout of mitoNEET in B-cell acute lymphoblastic leukemia affects bioenergetics

• KO cells have reduced oxidative phosphorylation and increased ROS production

• KO cells show reduced lipid content with reduction in cell proliferation

• KO cells exhibit increased sensitivity to cytarabine (Aca-C)