Targeting the leukemia cell metabolism by the CPT1a inhibition: functional preclinical effects in leukemias

Scientific evidence shows that the aberrant activation of signal transduction cascades is directly involved in the acquisition of new metabolic characteristics by tumor cells.^1-9  Recent studies have focused on the key role of fatty acid oxidation (FAO), as a way to produce adenosine triphosphate, reduced nicotinamide adenine dinucleotide phosphate, and acetyl–coenzyme A in neoplastic cells.¹⁰  The carnitine palmitoyl transferase 1a (CPT1a) catalyzes the rate-limiting step for the entry of fatty acids into the mitochondria, by loading fatty acyl-groups onto carnitine. It has been previously demonstrated that CPT1a interacts with members of the apoptotic machinery, such as Bcl-2 and t-Bid, and its inhibition can cause an accumulation of the toxic metabolite palmitate, resulting in mitochondrial damage and cell death.^11,12  Inhibition of CPT1a has already been applied in cancer research with positive results. ST1326 is an agent characterized by a reversible inhibitory activity toward the liver isoform of the enzyme CPT1a. ST1326 was initially selected by Sigma-Tau laboratories as a candidate for diabetes and ketoacidosis therapy.^13,14  Results reported by Pacilli et al¹⁵  demonstrated that the pharmacologic inhibition of FAO by ST1326 impairs cancer cell survival and inhibits tumor cell proliferation in in vitro and in vivo models of Burkitt lymphoma. Therefore, in this study we aimed to expand this approach, investigating the in vitro antileukemic activity of ST1326 on human leukemia cell lines and primary cells obtained from patients with different hematologic diseases.

Details are reported in supplemental Materials and methods (see the Blood Web site).

Human normal bone marrow (BM) derived CD34⁺ cells were purchased from StemCell Technologies. Peripheral blood and/or BM aspirates samples were obtained from 31 leukemia patients (supplemental Table 1), who provided written informed consent, in accordance with the Helsinki Declaration and approved by the Institutional Review Board of the Sapienza University of Rome. Cells were suspended in complete medium to a starting concentration of 1.0 × 10⁶/mL and exposed to ST1326 (1 to 50 μM) and, for selected experiments, to etomoxir (50 to 200 μM). Details of cell culture of combination experiments are reported in supplemental Materials and methods.

Cell cycle distribution, annexin V (Ann V) binding, and mitochondrial transmembrane potential (ΔΨ_m) were performed as previously published.^16-18 

FAO rates were measured in real time by the XF24 Extracellular Flux Analyzer (Seahorse Bioscience). Cells were suspended in unbuffered Dulbecco’s modified Eagle medium supplemented with 2.5 mM glucose and 0.5 mM l-carnitine (both from Sigma-Aldrich) adjusted to pH 7.35, transferred in CellTak (BD Bioscience) coated plates, and incubated for 30 minutes at 37°C in a CO₂-free incubator. ST1326 or etomoxir (Sigma-Aldrich) were added 15 minutes before the assay to the indicated concentrations. Bovine serum albumin alone or conjugated to palmitate (Seahorse Bioscience) were added to each well just prior the assay, to a final concentration of 33 and 200 µM, respectively. Oxygen consumption rates were measured for basal state and following the sequential injection of oligomycin (1 µM), carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP) (0.4-0.8 µM), and a mix of antimycin A (1 µM) and rotenone (1 µM) (all from Seahorse Bioscience) in each well. After the assay, cells were detached and manually counted to assess cell viability.

Details are reported in supplemental Materials and methods.

We first demonstrated that all leukemia cell lines constitutively expressed CPT1a (Figure 1A) and that this enzyme activity was significantly inhibited by 2 μM ST1326 (supplemental Results and supplemental Table 2). Primary acute myeloid leukemia (AML) samples similarly to AML cell models express CPT1a: the mean densitometry values of CPT1a/COXIV ratio were 1.18 ± 0.12 and 1.12 ± 0.06, respectively (data not shown).

We next documented that ST1326 was effective in inducing dose-dependent FAO inhibition as demonstrated by real-time metabolic analysis in cell lines (Figure 1B) and primary samples (supplemental Figure 1). Conversely, 50 μM etomoxir induced only minor effects (supplemental Figure 2).

We then analyzed the activity of ST1326 in 3 different leukemia models demonstrating that the CPT1a inhibitor induced dose- and time-dependent cell growth arrest, cell cycle modulation, and apoptosis (Figure 1C-D and supplemental Figure 3). In particular, a marked depletion of the S phase, associated with an increase in G2M phase and sub-G0/1 peak, was documented after 24 hours of exposure to 50 μM ST1326. The same effects occurred also at low ST1326 concentrations over longer times of exposure (Figure 1D). The cytotoxic activity of the molecule was also confirmed by using Ann V (supplemental Figure 3). Etomoxir failed to reproduce similar significant results (supplemental Figure 3). By monitoring the retention of chloromethyl-X-rosamine (CMXRos), we demonstrated that ST1326 induced a significant reduction of ΔΨ_m (supplemental Figure 4). The loss of ΔΨ_m preceded the phospholipid exposure and the formation of sub-G0/1 peak on DNA histogram demonstrating that ST1326 induced apoptosis through activation of the mitochondrial pathway. Drug washout experiments demonstrating the irreversible cytotoxic activity of ST1326 are reported in supplemental Results and supplemental Figure 5.

To investigate whether lipid accumulation is part of the mechanism that induces apoptosis in leukemia cells, we used BODIPY-C16 to stain intracellular palmitate accumulation. A representative microphotograph in Figure 1E showed an increase in green fluorescence by accumulated palmitate when cells were treated with 10 µM ST1326. The quantification of the fluorescence intensities in all cells revealed a significant block of palmitate metabolism (Figure 1F), which reinforces the idea that ST1326 can reduce β-oxidation in leukemia cells.

The results obtained from the analysis of AML primary samples indicated that ST1326 exerts a dose- and time-dependent cytotoxic activity in all (17/17) samples from de novo patients (Figure 2). Induction of apoptosis was associated with a significant reduction of cell counts (Figure 2). Activity of ST1326 was confirmed by viability/carboxyfluorescein diacetate succinimidyl ester (CFSE) staining/CD34⁺ leukemic cells (supplemental Figure 6). ST1326 induced a significant increase in apoptosis also in samples from 2 relapsed AML patients: Ann V+ cells increased from 21.18% ± 7.75% to 25.23% ± 6.53% (P = not significant), 30.20% ± 3.27% (P = not significant), and 73.09% ± 4.11% (P = .00018) in the presence of 10, 20, and 50 μM ST1326, respectively (data not shown). Data demonstrating the favorable interaction of ST1326 with cytarabine (AraC) in AML cell lines and primary samples are reported in supplemental Results and supplemental Figure 7.

A time- and dose-dependent increase in the levels of apoptosis was observed in 5/5 samples from B-ALL and in 6/6 samples from chronic lymphoblastic leukemia analyzed after exposure to ST1326 (Figure 2). The apoptosis increase was associated with a significant (at 50 μM) cell count reduction (Figure 2). The sample obtained from a relapsed B-ALL patient was equally sensitive to ST1326 (data not shown).

In summary, our results show that there is a variability within and among leukemias in sensitivity to ST1326, which is not unexpected given the heterogeneity of these leukemias. These observations may support the perspective of leukemia patient stratification according to metabolic phenotypes.

No cytotoxic effects were detected on normal BM CD34⁺ cells as reported in the supplemental Results.

So far, lipid metabolism in cancer cells has been investigated mainly from the anabolic perspective, as fatty acid synthase is overexpressed in many cancers (reviewed in Menendez and Lupu¹⁹ ). Conversely, the importance of FAO in cancer cell bioenergetic balance has been poorly established. Our results agree with the data previously reported by Samudio et al,²⁰  reinforcing the idea that β-oxidation represents a crucial metabolic pathway for leukemic cells, not only for energy production but also for controlling cell growth, survival, and chemoresistance. Moreover, these observations have been strengthened by subsequent works on other hematologic malignancies.^21-23 

Taken together, these findings emphasize the roles of FAO as a potential target for leukemia treatment and of ST1326 as a promising agent able to effectively inhibit FAO on leukemia cells, leading to cell growth reduction and apoptosis.

The authors thank RomaAIL (Associazione Italiana contro le Leucemie-linfomi e mieloma) for providing laboratory space and D. Calef for editing the manuscript.

This work was supported by grants from Sapienza (C26G13WSNF and C26A137XXS), Ministero dell'Istruzione, dell'Università e della Ricerca (PRIN 2010AX2JX7_003), Fondi strutturali europei “Ricerca e Competitività” (PON01_01802; PON01_02512), Sigma-Tau S.p.A., and Fondazione Internazionale D’AMATO Onlus.

Contribution: M.R.R. designed the research, performed experiments, analyzed the data, and wrote the manuscript; S.M. conducted and analyzed FAO assays; R.L. and M.A. performed functional experiments and assisted with cell culture; A.C. performed CPT1a expression and activity experiments and fluorescence assays; R.F. provided samples and clinical data; M.R.T. provided critical revision of the manuscript; R.N. and G.P. provided scientific advice and critical revision of the manuscript; and A.T. designed the research, supervised the project, and wrote the manuscript.

Conflict-of-interest disclosure: R.N. is employed at Sigma-Tau S.p.A. as medical research senior advisor. Sigma-Tau is an Italian pharmaceutical company owner of the patent for ST1326. The coauthor declares that there are no conflicts of interest regarding the manuscript because she did not receive any financial supports or benefits, nor did any member of her immediate family, for writing this paper. The remaining authors declare no competing financial interests.

Correspondence: Agostino Tafuri, Hematology, “Sant’Andrea” Hospital–Sapienza, University of Rome, Department of Clinical and Molecular Medicine, Via di Grottarossa 1035, 00189 Rome, Italy; e-mail: agostino.tafuri@uniroma1.it.