Noncanonical TCA cycle fosters canonical TCA cycle and mitochondrial integrity in acute myeloid leukemia

Abstract

Cancer cells rely on mitochondrial oxidative phosphorylation (OXPHOS) and the noncanonical tricarboxylic acid (TCA) cycle. In this paper, we shed light on the vital role played by the noncanonical TCA cycle in a host‐side concession to mitochondria, especially in highly energy‐demanding malignant tumor cells. Inhibition of ATP‐citrate lyase (ACLY), a key enzyme in the noncanonical TCA cycle, induced apoptosis by increasing reactive oxygen species levels and DNA damage while reducing mitochondrial membrane potential. The mitochondrial membrane citrate transporter inhibitor, CTPI2, synergistically enhanced these effects. ACLY inhibition reduced cytosolic citrate levels and CTPI2 lowered ACLY activity, suggesting that the noncanonical TCA cycle is sustained by a positive feedback mechanism. These inhibitions impaired ATP production, particularly through OXPHOS. Metabolomic analysis of mitochondrial and cytosolic fractions revealed reduced levels of glutathione pathway‐related and TCA cycle‐related metabolite, except fumarate, in mitochondria following noncanonical TCA cycle inhibition. Despite the efficient energy supply to the cell by mitochondria, this symbiosis poses challenges related to reactive oxygen species and mitochondrial maintenance. In conclusion, the noncanonical TCA cycle is indispensable for the canonical TCA cycle and mitochondrial integrity, contributing to mitochondrial domestication.

Abbreviations

• 2DG
• 2-deoxy-d-glucose
• ACLY
• ATP-citrate lyase
• GMFI
• geometric mean fluorescence intensity
• MES
• 2-[N-morpholino] ethanesulfonic acid
• MMP
• mitochondrial membrane potential
• NAC
• N-acetyl-l-cysteine
• OligA
• oligomycin A
• OXPHOS
• oxidative phosphorylation
• ROS
• reactive oxygen species
• TCA
• tricarboxylic acid
• TMRE
• tetramethylrhodamine ethyl ester

INTRODUCTION

Approximately 2 billion years ago, the incorporation of mitochondria into eukaryotic cells enabled efficient energy production.^1–4 While benefiting from this symbiotic relationship, the host organism required advancements in their defense mechanisms against ROS that are generated at high levels during OXPHOS and in metabolic pathways to coexist symbiotically with mitochondria.

ACLY (EC 4.1.3.8) is a key enzyme of the noncanonical TCA cycle in mitochondria^5,6 (Figure 1A). As acetyl-CoA cannot directly penetrate the mitochondrial membrane, it is converted to citrate within mitochondria by the action of citrate synthetase (CS). While citrate is incorporated into the canonical TCA cycle, it is also exported from mitochondria to the cytoplasm through the mitochondrial membrane citrate/malate transporter, SLC25A1. Subsequently, ACLY regenerates acetyl-CoA and oxaloacetate from cytosolic citrate, thereby providing acetyl-CoA in the cytosol as a substrate for lipid synthetase⁷ and histone acetylation.⁸ Cytosolic oxaloacetate is then converted to malate, which can be transported back into mitochondria through SLC25A1 in the noncanonical TCA cycle. Therefore, citrate serves as a crucial substrate that links the canonical and the noncanonical TCA cycles.

[IMAGE OMITTED. SEE PDF]

Metabolic reprogramming is one of the hallmarks of cancer.^9–11 OXPHOS within mitochondria supports robust tumor growth, providing energy through the canonical TCA cycle.^10,12 The canonical TCA cycle has been extensively explored in cancer metabolism,^9,10,12 however, the biological role of the noncanonical TCA cycle¹³ is yet to be fully elucidated. Of note, ACLY expression is elevated in a variety of cancers, including prostate cancer,¹⁴ liver cancer,¹⁵ non–small-cell lung carcinoma,¹⁶ colon cancer,¹⁷ and breast cancer.¹⁸ In these cancers, noncanonical TCA cycle inhibition using ACLY inhibitors induces in vitro antitumor effects.^5,19–21 In metabolism-associated fatty liver disease and steatohepatitis, ACLY targeting treatments have been developed to inhibit fatty acid synthesis.^7,22,23 Phase III clinical trials on an ACLY inhibitor, bempedoic acid, have demonstrated favorable outcomes for hyperlipidemia treatment, with good safety and tolerability profiles.²⁴ Thus, targeting the noncanonical TCA cycle could represent a novel strategy for cancer treatment.

In this study, we explored metabolic interactions between the noncanonical and the canonical TCA cycles, revealing that ACLY plays an essential role in supporting mitochondrial function and the canonical TCA cycle.

MATERIALS AND METHODS

Maintenance of cell lines

The original source of a fresh AML cell line (ATCC, VA, USA) was divided into vials and cryopreserved at −180°C using the CELLBANKER cell freezing medium (AMBSIO, OX, UK). After thawing, the cells were maintained and passaged for up to 2 months, or until morphological changes were observed. The cells were cultured in RPMI1640 medium (Biowest, Nuaillé, France) supplemented with 10% heat-inactivated FBS, penicillin (100 U/mL), and streptomycin (100 mg/mL) at 37°C in a 5% CO₂ incubator. The medium was prepared and cryopreserved at −20°C and thawed overnight at 4°C; then, it was renewed every 2–3 days at a cell density between 1 × 10⁵ and 1 × 10⁶ cells/mL. Cultured cells were screened for mycoplasma infection.

Drug sensitivity assay

To determine the drug sensitivity, a CellTiterBlue (CTB) assay was performed according to the instructions of the manufacturer (Promega, Madison, WI, USA). Briefly, cells were seeded in triplicate in 384-well transparent plates with flat bottoms (5 × 10³ cells/well), and medium with or without seven serially diluted reagent concentrations was added to the wells. After 68 h, the cells were incubated for another 4 h with the CTB reagent (Promega, WI, USA, G8081). Absorbance at 570 nm was measured using a TECAN Spark® microplate spectrophotometer (TECAN, Männedorf, Switzerland), with 600 nm used as the reference wavelength. The reagents used in this study are listed in Table S5.

Real-time RT-PCR analysis

Gene expression levels were quantified using real-time RT-PCR with SYBR Green I following the comparative ΔΔCT method.²⁵ RNA was extracted using an RNeasy mini kit (Qiagen, Hilden, Germany) and converted into cDNA using the EvoScript Universal cDNA Master Kit (Roche, Manheim, Germany). The primer pairs used in this study are listed in Table S6. ACTB or GAPDH expression was used as the internal control.

Western blot analysis

Cell lysate extracted using RIPA buffer was denatured in Laemmli sample buffer (Genscript, NJ, USA) and mercaptoethanol for 5 min at 95°C. Samples containing 10 μg of protein, measured using the bicinchoninic acid method, were loaded onto an 8% or 4%–20% gradient SurePAGE precast gel (Genscript) and electrophoresed at 200 V for 25 min in MES running buffer. Proteins were transferred onto PVDF membranes (Bio-Rad, Hercules, CA, USA) at 30 V for 12 h or overnight at 4°C in a transfer buffer containing 20% methanol. After the transfer, PVDF membranes were blocked for 1 h using 5% bovine serum albumin (lyophilized, pH ~7.0) (Atlantis Bioscience, Singapore, P6154-100G) and incubated with primary antibodies at 4°C for 12 h or overnight. After incubation with secondary antibodies, the membranes were agitated in Clarity and Clarity Max ECL western blotting substrates (Bio-Rad), and luminescence was detected using Bio-Rad ChemiDoc/Imaging system (Bio-Rad). The antibodies used for this analysis are listed in Table S7. Densitometric analysis was performed with ImageJ software version 1.53t (National Institutes of Health, Bethesda, MD, USA).

ACLY and CS enzymatic activity assays, and quantitation of citrate and acetyl-CoA

ACLY activity was assessed using the ACLY assay kit (BPS Bioscience, CA, USA, #79904). Cell-free analysis of ACLY enzymatic activity was performed in the absence or presence of five concentrations (7.8–250 μM) of ACLY inhibitor or CTPI2 (MedChemExpress, South Brunswick, NJ, USA; HY-123986). Intracellular ACLY activity was determined using cell lysates prepared in ACLY buffer as previously described.²⁶ Intracellular CS activity was also measured in cell lysates using a Citrate Synthase Activity Assay Kit (Abcam, ab119692). Intracellular citrate and acetyl-CoA levels were quantified using Citrate assay (Abcam, ab83396) and Acetyl-CoA assay (MyBioSource, Inc., CA, USA, MBS9719208) kits, respectively. All assays were performed following the manufacturer's instructions using the TECAN a Spark® microplate spectrophotometer (TECAN) and normalized to the total protein levels in lysates, quantified using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA, #23227).

Flow cytometric analysis

Flow cytometric analysis was performed using a BD LSR II flow cytometer and a FACS Celesta cell analyzer (BD Biosciences, MA, USA). Apoptotic events were assessed after staining with anti-Annexin-V antibodies conjugated with AF647 or Hoechst 33342. Levels of mitochondrial ROS, specifically superoxide, MMP, were measured using MitoSOX (Thermo Fisher Scientific, M36008) and TMRE (MedChemExpress, HY-D0985A), respectively. Data were analyzed using FlowJo version 10.9.0 software (BD Biosciences). To ameliorate ROS-induced damage, cells were suspended in RPMI 1640 medium supplemented with 10% FBS, and 3 mM NAC (Merck, Darmstadt, Germany, A7250) and ACLY inhibitor. Analyses were performed 12 h after the addition of ACLY inhibitor, at which time apoptosis was evident.

Energy-shift analysis under noncanonical TCA cycle inhibition

We further investigated the energy-shift by ACLY inhibition developing the ATP Tracking Harmonized ENergy-shift Assay (ATHENA). For ATHENA, the culture medium was changed the night before the experiment, and ACLY inhibitor and/or CTPI2 were added to 1 × 10⁶ cells in 2 mL of medium. After 6 h of culture, the cell-containing medium was divided into four wells for the control, 1 μM OligA (MedChemExpress, #75351), 100 mM 2DG (Merck, D6134), and OligA and 2DG combination treatments. Subsequently, 2DG and OligA were added at 0 and 10 min, respectively, because OligA acts faster than 2DG. Biotracker ATP-Red Live Cell Dye (Merck, SCT045)²⁷ was added to the medium at a final concentration of 10 μM and incubated at 37°C in a 5% CO₂ incubator for 15 min for direct ATP labeling. After washing in cold PBS, the cell pellet obtained was resuspended in 200 μL cold PBS containing Hoechst 33342 at a final concentration of 1 μg/mL. Treated cells were immediately analyzed using a phycoerythrin channel equipped with a yellow-green laser through FACS LSRII. Dependency on OXPHOS and glycolysis was calculated as the impaired ATP production rate (%) based on OligA and 2DG, which suppressed ATP production, using the following formula:$\text{Impaired}\mathrm{ATP}\text{production rate}\left(\%\right)=\frac{{\overline{X}}_{\left(\text{ctrl}\right)}-X_{\mathrm{i}}}{{\overline{X}}_{\left(\text{ctrl}\right)}-{\overline{X}}_{\left(\text{bottom}\right)}}\times 100$

Where X represents the GMFI value for each condition. X_i: GMFI for the condition of interest; X_(ctrl): GMFI without treatment with OligA or 2DG in the absence of the ACLY inhibitor and CTPI2; X_(bottom): GMFI for treatment with OligA and 2DG following culture with the ACLY inhibitor in combination with CTPI2; based on these values, we calculated the impaired ATP production rate under each condition.

Metabolomic analysis using LC-MS

THP-1 cells were expanded to 5 × 10⁷ cells per treatment and maintained at a maximum concentration of 5 × 10⁵ cells/mL in RPMI 1640 medium supplemented with 10% FBS. Then, the cells were treated with ACLY inhibitor and/or CTPI2 for 6 h. Subsequently, they were fractionated into mitochondrial and cytosolic components. In brief, harvested THP-1 cells were resuspended in 1 mL of ice-cold PBS and then centrifuged at 200 × g for 3 min. Next, the supernatant was discarded, and cells were resuspended in 300 μL of sucrose cell extraction buffer (SCEB; 300 mM sucrose, 10 mM HEPES at pH 7.4, 50 mM KCl, 5 mM EGTA, and 5 mM MgCl₂). The cells were equilibrated on ice for 30 min and then lysed by passage through a 25G needle 25 times. The supernatant containing intact mitochondria and cytosolic lysate was transferred to a new microtube. Unbroken cell pellets were resuspended in 100 μL SCEB and disrupted by additional passage through a 30G needle 10 times. The cell lysate was centrifuged at 2000 × g for 3 min at 4°C to pellet nuclei and intact cells. The supernatants were pooled, and the pellets were again passed through a 30G needle. This step was repeated twice. Then, the pooled supernatants were centrifuged at 7000 × g for 10 min at 4°C to pellet whole mitochondria. The supernatant, which was composed of the cytosolic fraction, was transferred to a new tube. The mitochondrial pellet was washed with SCEB and centrifuged at 7000 × g for 10 min at 4°C and was lysed in 100 μL RIPA buffer on ice for 30 min. Protein concentrations were determined through the bicinchoninic acid assay for the normalization of subsequent metabolite measurements.

Mitochondrial and cytosolic fractions were adjusted to 250 μL by adding water. Subsequently, 500 μL methanol and 500 μL chloroform were added to the samples, followed by vortexing for 5 min, and the mixture was centrifuged at 21,500 × g for 10 min. After centrifugation, the entire upper layer was passed through an Amicon Ultra 0.5-mL centrifugal filter (3 kDa MWCO, Merck). The filtrate was dried using a centrifugal vacuum concentrator (TAITEC Corporation, Saitama, Japan). The resultant components were dissolved in 200 μL water, and the diluted supernatant was stored in a LabTotal vial (Shimadzu Corporation, Kyoto, Japan). The analysis was performed using a high-pressure LC-equipped LC-MS-8060 (Shimadzu Corporation) system. For analysis of primary metabolites, the LC/MS/MS method package for primary metabolites version 3 was used. The ultra-high-performance liquid chromatographic conditions were as follows: Discovery HS F5-3 column (150 × 2.1 mm I.D., 3 μm particle size; Merck) with mobile phases A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile) employed for LC separation. The gradient for mobile phase B was programmed as follows: 0% (2 min)–25% (5 min)–35% (11 min)–50% (12.5 min)–50% (16 min)–95% (17 min). The column oven temperature was 40°C. Data processing was performed using LabSolutions software (version 5.82 SP1; Shimadzu Corporation). The area under the curve value for each metabolite was normalized to the protein concentration and transformed to a logarithmic scale (base 10). Subsequently, the values were converted to relative values against those for the no-treatment control for both mitochondrial and cytosolic fractions. A mean-centering method was applied for heatmap construction by dividing each variable by its standard deviation. Hierarchical clustering was performed using Ward's method based on Euclidean distances. MetaboAnalyst version 6.0 () was used for the analysis of metabolomic data.

Statistical and bioinformatic analysis

All data were expressed as the median and standard deviation of the results of three independent experiments. Unpaired Student's t-test, one-way ANOVA with Tukey's post hoc test, and two-way ANOVA with Dunnett's post hoc test were performed using GraphPad Prism version 10.0.2 (GraphPad Software, CA, USA). Statistical significance was set at p < 0.05. The levels of significance are indicated as follows: *0.01 ≤ p < 0.05; **0.001 ≤ p < 0.01; ***0.0001 ≤ p < 0.001; ****p < 0.0001. The synergistic and antagonistic effects of reagents were determined, based on cell viability, as zero-interaction potency (ZIP) score using the SynergyFinder version 3.0.1²⁸ in R software version 4.2.0 (R Foundation for Statistical Computing, Vienna, Austria). Publicly available clinical data on ACLY expression and patient prognosis were assessed using PedSCAtlas (),²⁹ based on the bulk RNA-seq TARGET study database. The log-rank test was performed using the cut-off score and median ACLY expression in the AML patient cohort.

RESULTS

ACLY contributes to AML cell survival and its inhibition exerts an antileukemic effect

To analyze the prognostic value of ACLY in AML patients, we analyzed bulk RNA-sequencing data and clinical information from the TARGET initiative database () using PedSCAtlas.³⁰ Patients with high ACLY expression subgroup exhibited a significantly poorer prognosis than those with low ACLY expression (Figure 1B). The same tendency was reported in three other independent AML clinical cohorts in Belgium, the Netherlands, and China, as well as in TCGA database from the USA.³¹ However, ACLY inhibitor did not exhibit any synergistic effect with conventional AML chemo-reagents (cytarabine, doxorubicin, and etoposide) in the THP-1 cell line (Figure S1A,B).

To elucidate the role of ACLY in cell proliferation and to explore the potential of ACLY inhibition for AML treatment, we assessed sensitivity to ACLY inhibitor, BMS-303414 (Cayman Chemical, MI, USA, #16239), using the THP-1, U937, and MOLM13 AML cell lines using CTB reagent (Figure 1C). The three cell lines exhibited similar sensitivities; therefore, we focused on the THP-1 AML cell line,³² a classical AML cell line with the highest in vitro sensitivity to ACLY inhibitor. We also performed a time-course assessment on the number of live cells at each concentration of ACLY inhibitor (Figure 1D). We observed that 70 μM ACLY inhibitor induced cell death and inhibited THP-1 cell growth, particularly after 48 h of culture.

The expression of integrative stress response genes, including PKR, GCN2, and ATF4, also increased following ACLY inhibition (Figure S2A), suggesting that ACLY inhibition induced various types of stress, including nutrient depletion and/or ROS-induced stress.

ACLY inhibition induces apoptosis in the THP-1 cell line

Next, we investigated the mechanism of the antileukemic effect of ACLY inhibition. After 6 h of culture, 70 μM ACLY inhibitor induced early apoptosis, and this effect was more apparent after 12 h of culture (Figure 2A,B). Leishman–Giemsa staining revealed blebbing and vacuolation, which are early signs of apoptosis, in cells treated with 70 μM ACLY inhibitor for 12 h (Figure 2C). Western blotting analysis revealed an increase in γH2AX, PARP and caspase 3 cleavage between 6 and 12 h of culture with 70 μM ACLY inhibitor (Figure 2D), confirming the induction of DNA damage and apoptosis.

[IMAGE OMITTED. SEE PDF]

ACLY inhibition increases mitochondrial ROS levels and compromises mitochondrial integrity

Based on the hypothesis that the noncanonical TCA cycle supports mitochondrial function, we explored the involvement of ROS in apoptosis by ACLY inhibition. First, we performed a time-course flow cytometric analysis of mitochondrial ROS levels in THP-1 cells (Figure 3A). The cell population with high mitochondrial ROS levels consistently increased after 3 h of culture with 70 μM ACLY inhibitor (Figure 3B). Additionally, the expression of genes related to the antioxidant system, such as GSS, NRF2, and TRX2, was upregulated following ACLY inhibitor treatment (Figure S2B). Of note, ROS induction upon ACLY inhibition preceded early apoptosis (Figure 2A). Next, we assessed MMP impairment upon ACLY inhibition. The low-MMP fraction was determined as the range from the unstained negative control in THP-1 cells not subjected to ACLY inhibitor treatment in the histogram (Figure 3C). After 6 h of culture with 70 μM ACLY inhibitor, the population of cells with low MMP (Figure 3D) was increased. To determine whether ROS mediated ACLY inhibition-induced MMP impairment, THP-1 cells were simultaneously incubated with ACLY inhibitor and 3 mM NAC. MMP impairment and apoptosis by ACLY inhibitor were alleviated by supplementation with 3 mM NAC (Figure 3E,F). Collectively, ACLY inhibition induced apoptosis through intracellular ROS accumulation, decreased MMP, and compromised mitochondrial integrity.

[IMAGE OMITTED. SEE PDF]

ACLY inhibition increases dependency on OXPHOS and reduces ATP synthesis as determined using the ATHENA method

To assess the metabolic-profile alteration following ACLY inhibition, first, we assessed the intracellular ACLY enzymatic activity after 6 h ACLY inhibition in THP-1 cells. ACLY inhibition impaired intracellular ACLY activity (Figure 4A), enhanced intracellular CS activity (Figure 4B), and decreased intracellular citrate amounts (Figure 4C) in a dose-dependent manner. Intriguingly, intracellular acetyl-CoA amount was maintained even though ACLY activity decreased (Figure 4D), and, ATP level was reduced after 70 μM ACLY inhibitor treatment (Figure 4E,F).

[IMAGE OMITTED. SEE PDF]

We further analyzed the pathway-dependent ATP production under conditions of OXPHOS and glycolysis suppression after ACLY inhibition using the ATHENA method. First, in the absence of ACLY inhibitor, intracellular ATP level was more decreased after OXPHOS inhibition (GMFI rate to the no-treatment control: 77%) than that after glycolysis inhibition (93%) (Figure 4G). This result indicated ATP dependency on OXPHOS over glycolysis in THP-1 cells. However, after culturing cells with ACLY inhibitor, basal ATP levels and concentrations in live cells decreased (77% and 38% with 35 and 70 μM ACLY inhibitor, respectively). Moreover, ATP production was more impaired following ACLY inhibition under OXPHOS inhibition (impaired ATP production rate: 54%, 97%, and 78% with 0, 35, and 70 μM ACLY inhibitor, respectively) than under glycolysis inhibition (16%, 26%, and 12% with 0, 35, and 70 μM ACLY inhibitor, respectively) (Figure 4H). Inhibition of the ACLY pathway also tended to further decrease ATP synthesis via OXPHOS compared with glycolysis in U937 and MOLM13 cells (Figure S3D). These results indicated that the noncanonical TCA cycle supported the ATP production from OXPHOS in AML cells.

The combined ACLY inhibitor and CTPI2 treatment exerts a potent synergistic antitumor effect through noncanonical TCA cycle inhibition

To elucidate the involvement of the noncanonical TCA cycle in supporting the canonical TCA cycle, we investigated the link between two representative noncanonical TCA cycle components through the combination treatment with ACLY inhibitor and mitochondrial citrate transporter inhibitor, CTPI2 (Figure 1A). First, the IC₅₀ values for the ACLY inhibitor and CTPI2 were 44.9 and 77.1 μM, respectively. (Figure 5A). Subsequently, we selected their highest non-toxic concentrations based on their dose–response curves (Figure 5A). Hereafter, we refer to this most synergistic noncanonical TCA cycle-suppressing combination (35 μM ACLY inhibitor and 60 μM CTPI2) (Figure 5B–D) as “the combination treatment” and to the apoptosis inducible treatment using 70 μM ACLY inhibitor as “the intensive ACLY inhibitor treatment.” A significant synergistic effect was observed following the combination treatment for 72 h on ZIP score (Figure 5B,C). This unique synergistic effect between ACLY inhibitor and CTPI2 was also observed in two other AML cell lines, U937 and MOLM13 (Figure S3A–C). Interestingly, noncanonical TCA cycle inhibition with the combination treatment induced a high ROS fraction (Figure 5E) and decreased MMP (Figure 5F), similar to those with the intensive ACLY inhibitor treatment (Figure 3B,D). The combination treatment also induced integrative stress- and antioxidant system-related gene expression (Figure S4A,B), similarly to those with the intensive ACLY inhibitor treatment (Figure S2A,B). Furthermore, similar to the intensive ACLY inhibitor treatment (Figure 4A–C), the combination treatment suppressed the ACLY activity (Figure S5A), enhanced the CS activity (Figure S5C), and reduced intracellular citrate levels (Figure S5D). Both the combination and the intensive ACLY inhibitor treatments induced apoptosis and γH2AX production (Figures 2A–D and S4C,D). Surprisingly, we observed that CTPI2 suppressed ACLY activity (Figure S5A) after 6 h of treatment, although it did not suppress ACLY activity in the cell-free system (Figure S5B). The combination treatment did not alter acetyl-CoA levels (Figure S5E) but increased the low-ATP fraction (Figure 5G). Furthermore, as with the intensive ACLY inhibitor treatment (Figure 4H), the combination treatment increased ATP production impairment under OXPHOS inhibition conditions (impaired ATP production rate: 54% with no reagent use, and 97%, 98%, and 96% at 35 μM ACLY inhibitor, 60 μM CTPI2, and 35 μM ACLY inhibitor and 60 μM CTPI2 in combination, respectively) compared with the impairment under glycolysis inhibition (16%, 26%, 25%, and 30%, respectively) (Figure S5F,G). The tendency of the noncanonical TCA cycle depending on OXPHOS was also observed in U937 and MOLM13 cells (Figure S3E). In addition, NAC supplementation ameliorated apoptosis during the implementation of the combination treatment (Figure 5H). These findings indicated that the intensive ACLY inhibitor and the combination treatments inhibited the noncanonical TCA cycle in a similar manner, thereby inducing apoptosis.

[IMAGE OMITTED. SEE PDF]

Noncanonical TCA cycle directly supports mitochondrial energy production by the canonical TCA cycle as determined through the metabolomic analysis

Inhibition of the noncanonical TCA cycle reduced intracellular citrate levels (Figures 4C and S5D). Considering that citrate is the common substrate between the mitochondrial canonical and cytosolic noncanonical TCA cycles, we conducted a comprehensive metabolomic analysis to distinguish cytosolic and mitochondrial fractions. We assessed 102 metabolites, including TCA cycle-related metabolites. After excluding metabolites that showed no signal in the cytosol and mitochondria, 62 metabolites were selected for further analysis (Table S1). After normalization, we constructed a heatmap for the metabolomic assay (Figure 6A,B) and focused on the levels of metabolites, which consistently increased or decreased in the cytoplasm and mitochondria following implementation of the intensive ACLY inhibition or combination treatment (Tables S2 and S3). Thus, an alteration in the levels of selected metabolites was considered to reflect the burden on the noncanonical TCA cycle. Thus, in mitochondria, the levels of metabolites, including the TCA cycle metabolites, citrate, isocitrate, succinate, and malate, but not fumarate, decreased after 6 h of treatment, indicating a decrease in canonical TCA cycle activity in mitochondria (Figure 6A,B,F). The inhibition of the noncanonical TCA cycle induced a decrease in intracellular citrate levels (Figures 4C and S5D). Metabolomic analysis following subcellular fractionation revealed an apparent decrease in citrate levels in the mitochondrial fraction compared with its levels in the cytosolic fraction, particularly following the implementation of the combination treatment (Figure 6C,D). We also observed a significant decrease in the levels of several essential amino acids, including arginine, phenylalanine, asparagine, glutamine, and tryptophan, in mitochondria following the inhibition of the noncanonical TCA cycle (Figure 6B). Conversely, the levels of these essential amino acids increased in the cytosol (Figure 6A). We subsequently performed a pathway analysis with a focus on overlapping metabolites with commonly altered levels between the “intensive ACLY inhibitor treatment” and the “combination treatment” groups. The levels of metabolites involved in the glutathione pathways were decreased (Figure 6E and Table S4) after noncanonical TCA cycle inhibition. The mitochondrial oxidative/reductive glutathione ratio was increased after noncanonical TCA cycle inhibition. Furthermore, inhibition of the noncanonical TCA cycle reduced the levels of TCA cycle-related metabolites, except fumarate, in mitochondria, indicating impaired canonical TCA cycle activity (Figure 6F). As opposed to mitochondrial malate levels, which were decreased, cytoplasmic malate levels were maintained. There was no significant decrease in pyruvate level, which indicates the level of glycolysis, even under noncanonical TCA cycle inhibition. Collectively, these results indicated that inhibition of the noncanonical TCA cycle burdened the protective mechanisms against ROS by glutathione and disrupted the canonical TCA cycle and OXPHOS activity in mitochondria.

[IMAGE OMITTED. SEE PDF]

DISCUSSION

We elucidated the metabolic mechanism by which the noncanonical TCA cycle inhibition induces apoptosis, highlighting its critical role in maintaining mitochondrial integrity and energy production. Moderate inhibition of the noncanonical TCA cycle allowed for compensation through upregulated canonical TCA cycle activity. However, significant inhibition of the noncanonical TCA cycle through implementation of the intensive ACLY inhibitor or combination treatment reduced canonical TCA cycle-related metabolites in mitochondria. This was accompanied by increased ROS levels and DNA damage. The ATHENA method, inspired by the Seahorse technology,³³ and SCENITH method,³⁴ facilitated direct ATP measurement under OXPHOS and glycolysis suppression. This approach revealed a significant OXPHOS dependency for ATP production in the THP-1 cell line, particularly under the noncanonical TCA cycle inhibition (Figure 4G,H).

ACLY plays a pivotal role in fatty acid synthesis. However, we did not observe any decrease in the intracellular levels of fatty acids after inhibition of the noncanonical TCA cycle inhibition during 12-h culture (Figure S6A,B). These unexpected results indicated that extensive impairment of the noncanonical TCA cycle led to immediate energy depletion and apoptosis. Alteration in the fatty acid profile might be observed in surviving cells upon prolonged exposure to low ACLY concentrations. These results also underscore the fact that inhibition of the noncanonical TCA cycle does not cause cell death via the depletion of fatty acids in AML cells.

Although ACLY reduces the cytoplasmic conversion of citrate to acetyl-CoA and oxaloacetate, intracellular citrate levels were decreased following ACLY inhibition. Moreover, even though ACLY activity is expected to be upregulated when cytosolic citrate levels are decreased,^20,35,36 CTPI2, which inhibits citrate translocation from mitochondria to cytoplasm, impaired ACLY enzymatic activity. Notably, our cell-free analysis confirmed that CTPI2 did not directly impair ACLY activity (Figure S5B). These unexpected findings indicated that noncanonical TCA cycle inhibition at its intermediate step disrupts the upstream processes in a feedback loop, affecting ATP production, citrate levels, and ACLY activity. This was also validated by the apparent synergistic effect of the combination treatment (Figure 5A–C). Interestingly, the canonical TCA cycle was prioritized and maintained upon promotion of CS activity (Figures 4B,C and S5C,D) under moderate noncanonical TCA cycle inhibition. Collectively, these results indicated the critical role of the noncanonical TCA cycle in supporting the function of the canonical TCA cycle in mitochondria.

To distinguish between the levels of metabolites of the canonical and noncanonical TCA cycles, we performed an LC-MS analysis for each mitochondrial and cytosolic fraction. TCA cycle-related metabolites, except fumarate, were reduced in mitochondria (Figure 6F), which underscores the essential role of the noncanonical TCA cycle in maintaining the canonical TCA cycle. Fumarate inhibits succinate dehydrogenase, which constitutes complex II of the electron transport chain,^37,38 leading apoptosis.

Inhibition of the noncanonical TCA cycle was also associated with increased mitochondrial ROS levels, decreased MMP levels, and DNA damage in surviving cells, eventually resulting in apoptosis (Figures 2A–D and S3C,D). Although we observed several apoptotic cells microscopically after mild ACLY inhibition using 35 μM ACLY inhibitor (Figure 2C), the majority of surviving cells maintained the ability to remove mitochondrial ROS. In contrast, intensive ACLY inhibition using 70 μM ACLY inhibitor impaired ROS removal and decreased MMP even in surviving cells. In addition, NAC supplementation reversed the low MMP (Figures 3E and 5F) and apoptosis induced by noncanonical TCA cycle inhibition (Figures 3F and 5H). These results indicated that the inhibition of the noncanonical TCA cycle directly led to energy depletion and apoptosis via excessive mitochondrial ROS. Our metabolomics data showed reduced glutathione and significantly increased oxidative/reduced glutathione ratio in mitochondrial fraction following noncanonical TCA cycle inhibition. This suggested that the noncanonical TCA cycle plays a protective role against mitochondrial ROS-induced damage through the induction of the glutathione pathway (Figures 6E and S2B). Collectively, the noncanonical TCA cycle essentially contributes to the maintenance of mitochondrial integrity and function.

The existence of compensatory pathways for the noncanonical TCA cycle was speculated. In the glutamate pathway, glutamate dehydrogenase converts glutamate to α-ketoglutarate and provides the substrate for the canonical TCA cycle in mitochondria.^39,40 In addition, acetyl-CoA levels were maintained even under noncanonical TCA cycle inhibition (Figure 4D and S5D), suggesting a potential compensatory mechanism via acetyl-CoA production from acetate through the action of ACSS2 in the cytosol.^41,42 However, the two reagents, R162, a glutamate dehydrogenase inhibitor (Figure S7A–C), and an ACSS2 inhibitor (Figure S7D,E) exhibited mild synergistic effects with noncanonical TCA cycle inhibitors. Thus, the noncanonical TCA cycle plays a crucial role in supporting the canonical TCA cycle.

In summary, the noncanonical TCA cycle plays a pivotal role in fostering canonical TCA cycle and protecting against mitochondrial ROS damage. Considering this, it is plausible that ACLY participates in DNA damage repair⁴³ and in fatty acid synthesis by supplying acetyl-CoA⁷ for β-oxidation. The noncanonical TCA cycle represents a host-side concession for mitochondrial domestication and accommodation achieved in evolutionary steps. In particular, in highly energy-demanding malignant tumors, aberrant ACLY and the noncanonical TCA cycle activity confers a survival advantage. Therefore, ACLY and the noncanonical TCA cycle are promising therapeutic targets for malignant tumors.

AUTHOR CONTRIBUTIONS

Atsushi Watanabe: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; software; visualization; writing – original draft. Chartsiam Tipgomut: Methodology; resources; writing – review and editing. Haruhito Totani: Formal analysis; methodology; resources; writing – review and editing. Kentaro Yoshimura: Data curation; formal analysis; investigation; methodology; resources; software; writing – review and editing. Tomohiko Iwano: Data curation; formal analysis; investigation; methodology; resources; software; writing – review and editing. Hamed Bashiri: Writing – review and editing. Lee Hui Chua: Project administration; resources; writing – review and editing. Chong Yang: Methodology; writing – review and editing. Toshio Suda: Conceptualization; funding acquisition; project administration; resources; supervision; writing – review and editing.

ACKNOWLEDGMENTS

The authors thank the Cancer Science Institute of Singapore Core Facility members, Dr. Akiko Nambu (Cancer Science Institute of Singapore, National University of Singapore), Dr. Shih-Chia Yeh (Cancer Science Institute of Singapore, National University of Singapore), and Dr. Jianhong Ching (Duke-NUS medical college), for technical help and scientific advice.

FUNDING INFORMATION

This study was supported by the Singapore Translational Research Investigator Award from the National Medical Research Council of Singapore, NMRC/MOH-STaR18May (MOH-000149) to T.S.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT

All data and materials reported in this paper will be shared by the lead contact upon reasonable request.

ETHICS STATEMENT

Approval of the research protocol by an Institutional Reviewer Board: N/A.

Informed consent: N/A.

Registry and the registration no. of the study/trial: N/A.

Animal studies: N/A.

References

Gray MW, Burger G, Lang BF. Mitochondrial evolution. Science. 1999;283:1476‐1481.

Gray MW. Mitochondrial evolution. Cold Spring Harb Perspect Biol. 2012;4: [eLocator: a011403].

Margulis L. Origin of Eukaryotic Cells. Yale University Press; 1971.

Margulis L. Symbiosis in Cell Evolution. W.H. Freeman; 1981.

Icard P, Wu Z, Fournel L, Coquerel A, Lincet H, Alifano M. ATP citrate lyase: a central metabolic enzyme in cancer. Cancer Lett. 2020;471:125‐134.

Bauer DE, Hatzivassiliou G, Zhao F, Andreadis C, Thompson CB. ATP citrate lyase is an important component of cell growth and transformation. Oncogene. 2005;24:6314‐6322.

Morrow MR, Batchuluun B, Wu J, et al. Inhibition of ATP‐citrate lyase improves NASH, liver fibrosis, and dyslipidemia. Cell Metab. 2022;34:919‐936.e8.

Wellen KE, Hatzivassiliou G, Sachdeva UM, Bui TV, Cross JR, Thompson CB. ATP‐citrate lyase links cellular metabolism to histone acetylation. Science. 2009;324:1076‐1080.

Warburg O. On the origin of cancer cells. Science. 1956;123:309‐314.

Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science. 2009;324:1029‐1033.

Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646‐674.

Martínez‐Reyes I, Chandel NS. Mitochondrial TCA cycle metabolites control physiology and disease. Nat Commun. 2020;11:102.

Arnold PK, Jackson BT, Paras KI, et al. A non‐canonical tricarboxylic acid cycle underlies cellular identity. Nature. 2022;603:477‐481.

Shah S, Carriveau WJ, Li J, et al. Targeting ACLY sensitizes castration‐resistant prostate cancer cells to AR antagonism by impinging on an ACLY‐AMPK‐AR feedback mechanism. Oncotarget. 2016;7:43713‐43730.

Zhou Y, Tao J, Calvisi DF, Chen X. Role of lipogenesis rewiring in hepatocellular carcinoma. Semin Liver Dis. 2022;42:77‐86.

Migita T, Narita T, Nomura K, et al. ATP citrate lyase: activation and therapeutic implications in non‐small cell lung cancer. Cancer Res. 2008;68:8547‐8554.

Qiu Z, Deng W, Hong Y, et al. Biological behavior and lipid metabolism of colon cancer cells are regulated by a combination of sterol regulatory element‐binding protein 1 and ATP citrate lyase. Onco Targets Ther. 2021;14:1531‐1542.

Wang D, Yin L, Wei J, Yang Z, Jiang G. ATP citrate lyase is increased in human breast cancer, depletion of which promotes apoptosis. Tumour Biol. 2017;39: [eLocator: 1010428317698338].

Hatzivassiliou G, Zhao F, Bauer DE, et al. ATP citrate lyase inhibition can suppress tumor cell growth. Cancer Cell. 2005;8:311‐321.

Wei J, Leit S, Kuai J, et al. An allosteric mechanism for potent inhibition of human ATP‐citrate lyase. Nature. 2019;568:566‐570.

Lin R, Tao R, Gao X, et al. Acetylation stabilizes ATP‐citrate lyase to promote lipid biosynthesis and tumor growth. Mol Cell. 2013;51:506‐518.

Convertini P, Santarsiero A, Todisco S, et al. ACLY as a modulator of liver cell functions and its role in metabolic dysfunction‐associated steatohepatitis. J Transl Med. 2023;21:568.

Duarte Lau F, Giugliano RP. Adenosine triphosphate citrate lyase and fatty acid synthesis inhibition: a narrative review. JAMA Cardiol. 2023;8:879‐887.

Bays HE, Banach M, Catapano AL, et al. Bempedoic acid safety analysis: pooled data from four phase 3 clinical trials. J Clin Lipidol. 2020;14:649‐659.e6.

Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real‐time quantitative PCR and the 2(‐Delta Delta C(T)) method. Methods. 2001;25:402‐408.

Minic Z, Hüttmann N, Poolsup S, et al. Phosphoproteomic analysis of breast cancer‐derived small extracellular vesicles reveals disease‐specific phosphorylated enzymes. Biomedicine. 2022;10: [eLocator: 408]. doi:10.3390/biomedicines10020408

Wang L, Yuan L, Zeng X, et al. A multisite‐binding switchable fluorescent probe for monitoring mitochondrial ATP level fluctuation in live cells. Angew Chem Int Ed Engl. 2016;55:1773‐1776.

Ianevski A, Giri AK, Aittokallio T. SynergyFinder 3.0: an interactive analysis and consensus interpretation of multi‐drug synergies across multiple samples. Nucleic Acids Res. 2022;50:W739‐W743.

Mumme HL, Bhasin SS, Thomas BE, et al. Pediatric single cell cancer atlas: an integrative web‐based resource for single cell transcriptome data from pediatric leukemias. Blood. 2021;138:3488.

Mumme HL, Bhasin SS, Nawaz M, et al. A single cell atlas and interactive web‐resource of pediatric cancers and healthy bone marrow. Blood. 2022;140:2278‐2279.

Wang J, Ye W, Yan X, et al. Low expression of ACLY associates with favorable prognosis in acute myeloid leukemia. J Transl Med. 2019;17:149.

Aldo PB, Craveiro V, Guller S, Mor G. Effect of culture conditions on the phenotype of THP‐1 monocyte cell line. Am J Reprod Immunol. 2013;70:80‐86.

Chacko BK, Kramer PA, Ravi S, et al. The bioenergetic health index: a new concept in mitochondrial translational research. Clin Sci (Lond). 2014;127:367‐373.

Argüello RJ, Combes AJ, Char R, et al. SCENITH: a flow cytometry‐based method to functionally profile energy metabolism with single‐cell resolution. Cell Metab. 2020;32:1063‐1075.e7.

Bazilevsky GA, Affronti HC, Wei X, Campbell SL, Wellen KE, Marmorstein R. ATP‐citrate lyase multimerization is required for coenzyme—a substrate binding and catalysis. J Biol Chem. 2019;294:7259‐7268.

Verschueren KHG, Blanchet C, Felix J, et al. Structure of ATP citrate lyase and the origin of citrate synthase in the Krebs cycle. Nature. 2019;568:571‐575.

Tyrakis PA, Yurkovich ME, Sciacovelli M, et al. Fumarate hydratase loss causes combined respiratory chain defects. Cell Rep. 2017;21:1036‐1047.

Valcarcel‐Jimenez L, Frezza C. Fumarate hydratase (FH) and cancer: a paradigm of oncometabolism. Br J Cancer. 2023;129:1546‐1557.

Yang L, Venneti S, Nagrath D. Glutaminolysis: a hallmark of cancer metabolism. Annu Rev Biomed Eng. 2017;19:163‐194.

Li M, Li C, Allen A, Stanley CA, Smith TJ. The structure and allosteric regulation of mammalian glutamate dehydrogenase. Arch Biochem Biophys. 2012;519:69‐80.

Huang Z, Zhang M, Plec AA, et al. ACSS2 promotes systemic fat storage and utilization through selective regulation of genes involved in lipid metabolism. Proc Natl Acad Sci USA. 2018;115:E9499‐E9506.

Zhao S, Torres A, Henry RA, et al. ATP‐citrate lyase controls a glucose‐to‐acetate metabolic switch. Cell Rep. 2016;17:1037‐1052.

Sivanand S, Rhoades S, Jiang Q, et al. Nuclear acetyl‐CoA production by ACLY promotes homologous recombination. Mol Cell. 2017;67:252‐265.e6.