1 Introduction
Acute myeloid leukemia (AML) is a clonal malignant disease characterized by an uncontrolled proliferation of undifferentiated myeloid precursors, resulting in impaired normal hematopoiesis [
1]. Despite the emergence of novel treatment options approved for clinical practice in the past decades, the overall survival and relapse rates of AML patients have not been improved [
2]. Therefore, a novel and personalized therapy for AML patients needs to be developed.
The acyl-CoA synthetase long chain family member 5 (ACSL5), which belongs to the family of acyl-conenzyme A (CoA) synthetases (ACSs), is deeply involved in the regulation of fatty acid metabolism. ACS family members esterify free fatty acids to acyl-CoAs, which is a key activation step necessary for the utilization of fatty acids [
3,
4]. Comparing with other identified isoenzymes, ACSL5 activates specific substrates of long-chain fatty acids with a backbone of 12 to 22 carbons and prefers to catalyze palmitate, palmitoleate, oleate, and linoleate [
5]. ACSL5-knockout mice reported a 60% decrease in their palmitoyl-CoA synthesis rate [
6]. Apart from its role in energy metabolism, palmitoyl-CoA also serves as a palmitate donor for palmitoylation modification, which is a specific post-transcriptional modification of proteins. ACSL5 is also involved in regulating Wnt2b protein palmitoylation modification [
7]. Therefore, palmitoylated proteins can be regulated post-translationally by the activity of ACSL5 via the palmitoyl-CoA level. Several studies have reported that ACSL5 functions as a tumor promoter in malignant glioma [
8,
9] and a tumor suppressor in colorectal carcinomas [
10]. However, the biological and clinical significance of ACSL5 in AML has not yet been explored.
The Wnt family is involved in multiple developmental events during embryogenesis and has also been implicated in malignant disorders [
11]. The functions of Wnt proteins are regulated by different post-translational modifications, including esterification and glycosylation [
12,
13]. Glycosylation plays a critical role in protein folding and stability [
14], while esterification, specifically palmitoylation, is essential to Wnt secretion [
15]. The depalmitoylation of serine 209 of Wnt 3a protein could induce Wnt3a retention in the endoplasmic reticulum and failure in secretion [
16,
17].
Triacsin c, a pan-ACS inhibitor, could inhibit the enzyme activity of ACS family members, including ACSL5, by competing with fatty acids for their catalytic domain [
18]. This inhibitor has been reported to induce cytochrome c release and subsequent cell death via a functional apoptosome-mediated pathway, which can trigger antitumor activity in lung, colon, and brain cancer cells [
19]. However, the antitumor effect of triacsin c in AML has not yet been explored.
In this study, we investigated the prognostic significance of ACSL5 in cytogenetically normal AML (CN-AML) patients and further demonstrated ACSL5 as an oncogenic protein that enhanced cell growth and inhibited cell apoptosis. ACSL5 deficiency drastically suppressed the Wnt/β-catenin signaling pathway via the inhibition of Wnt3a palmitoylation. Moreover, triacsin c exhibited an antileukemia effect in both AML cell lines and primary cells and worked synergistically with ABT-199, the selective BCL-2 inhibitor. Therefore, we propose that ACSL5 could serve as a novel biomarker in prognosis prediction and a tantalizing therapeutic target of AML.
2 Materials and methods
2.1 Patients
Bone marrow specimens were collected from healthy donors and 312 CN-AML patients at the First Affiliated Hospital, Zhejiang University School of Medicine from July 2010 to April 2016. The clinical samples of the participants were obtained directly after their diagnosis and before initiating treatment. Routine laboratory and clinical parameters were recorded. The gene mutation profiles of the patients, including FLT3-ITD, NPM1, CEBPA, DNMT3A, and IDH1/2, were detected through PCR and sequencing. This study was approved by the ethics committee of the hospital. Written informed consents were obtained from all participants, and the Declaration of Helsinki was followed.
2.2 Cell lines and primary cells
The AML cell lines THP-1, KG-1, and OCI-AML2 were purchased from Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Shanghai, and THP-1-luciferase cell lines were donated by Professor Guido Marcucci of the City of Hope Medical Center and Beckman Research Institute, Duarte, California, USA. These cell lines were authenticated via DNA short-tandem repeat analysis by Shanghai Biowing Applied Biotechnology (Shanghai, China). THP-1, OCI-AML2, and KG-1 were routinely cultured in a Roswell Park Memorial Institute 1640 medium (RPMI-1640) supplemented with 10% fetal bovine serum (FBS) (Gibco). Primary AML cells were isolated via Ficoll-Hypaque (TBD Science, Tianjin, China) density gradient separation from the bone marrow samples of consenting patients who were newly diagnosed with AML. The primary cells were then cultured in a Stemspan serum-free medium II (SFEM II, StemCell Technologies) supplemented with recombinant human stem cell factor (IL-3, 25 ng/mL; IL-6, 10 ng/mL; Flt3L, 100 ng/mL; SCF, 50 ng/mL; TPO, 100 ng/mL). All cytokines were obtained from Pepro Tech US (Rocky Hill, NJ, USA). All cells were cultured at 37 °C with 5% CO2 and high humidity.
2.3 RNA extraction and quantitative real-time PCR
Total RNA was extracted with the RNAiso Plus reagent (Takara, Japan) followed by cDNA synthesis with PrimerScript RT Master Mix (Takara, Japan). A quantitative real-time PCR analysis for determining the mRNA level was performed by using the one-step SYBR PrimeScriptTM RT-qPCR Kit (Takara, Japan) operated on a IQ5 real-time PCR instrument (Bio-Rad, Hercules, CA). Relative RNA amount was calculated using the 2ΔΔCt method normalized to GAPDH. The primers were purchased from Sangon Biotech (Shanghai, China) and summarized as follows: GAPDH-Fwd: 5′-ATGGGGAAGGTGAAGGTCG-3′ and GAPDH-Rev 5′-CTCCACGACGTACTCAGCG-3′; ACSL5-Fwd: 5′-GGCATTGGTGCTGATAGG-3′; and ACSL5-Rev: 5′-TCTTCTCCCCTCTTTGCTT-3′.
2.4 Lentiviral preparation, viral infection, and stable cell lines generation
Lentiviral plasmids expressing shACSL5 (#1, #2) or shCtrl and ACSL5-overexpression (ACSL5-OE) or empty vector were purchased from Genecopoeia. Stably transduced cells were established following the procedure described in [
20]. To summarize, the indicated plasmids were introduced into HEK293T together with lentiviral packaging vectors psPAX2 and pMD2.G using Lipofectamine 2000 (Life Technologies, 12566014). Viruses were collected 48 h and 72 h after transfection, and the target cells (THP-1, OCI-AML2 and KG-1) were infected with the collected viruses in the presence of polybrene (Sigma, 107689). The infected cells were selected with puromycin (2 μg/mL; Thermo Fisher) 2 days after infection.
2.5 Cell growth and viability assay
Cell growth and viability assay were measured via the CellTiter 96 Aqueous One Solution cell proliferation assay (Promega). The infected cells (1 × 10
4 cells/mL) were seeded in 96-well plates and mixed with an MTS solution before being incubated for 4 h at 37 °C. Absorbance was detected at 490 nm daily for 7 consecutive days. AML cell lines (1 × 10
5 cells/mL) or primary cells were plated in 96-well plates and treated with the indicated concentrations of drugs. Drug synergy was analyzed using the Combenefit software (Loewe model) [
21]. The data were presented as means ± SEM from at least three independent experiments.
2.6 EdU assay
5-ethynyl-20-deoxyuridine (EdU) with Alexa-Fluor 594 dye kit (Beyotime, Shanghai, China) was used to measure cell proliferation ability. Cells were seeded with a density of 1 × 106 cells/mL and then incubated with a 10 μmol/L EdU buffer for 2 h at 37 °C, fixed with 4% formaldehyde for 15 min, and permeabilized with 0.3% Triton X-100 for 15 min. The results were assessed via flow cytometry.
2.7 Cell apoptosis and cell cycle assays
Cell apoptosis and cell cycle were analyzed via flow cytometry. For the apoptosis analysis, the cells were collected and stained with Annexin V-APC (or Annexin V-FITC) and PI at room temperature for 15 min. For the cell cycle analysis, the cells were harvested, fixed with 70% ethanol overnight at −20 °C prior to propidium iodide (PI) staining, and then analyzed immediately with a flow cytometer (NovoCyte, ACEA Biosciences). The data were analyzed using the NovoExpress software (ACEA Biosciences). Cell debris and dead cells were gated out for cell cycle analysis.
2.8 Western blot analysis
The cells were lysed in a RIPA buffer (Thermo Scientific, Waltham, MA) containing a protease and phosphatase inhibitor (Thermo Scientific, USA). Protein concentration was determined using the Bicinchoninic Acid Protein Assay Kit (Thermo Scientific, USA). The protein samples were separated by SDS-PAGE and transferred to 0.45 μm PVDF membranes (Millipore, USA). The membranes were blocked with 5% non-fat milk in TBST for 1 h, incubated overnight at 4 °C with primary antibodies, washed, and then incubated with HRP-linked specific secondary antibodies. Proteins were visualized with the ChemiDoc MP Imaging System (Bio-rad, Hercules, CA) using an enhanced chemiluminescence detection kit (Fdbio, China). GAPDH and β-tubulin were selected as loading controls. All antibodies employed in this study are listed below.
2.9 Antibodies and reagents
The antibodies for GAPDH (#5174), β-tubulin (#2146), PARP (#9532), and Caspase3 (#9662) were purchased from CST (MA, USA). The Wnt3a (ab219412), streptavidin-HRP (ab7403), and anti-human CD45-PE (ab27287) antibodies were purchased from Abcam (MA, USA), the anti-ACSL5 (15708-1-AP) antibody was purchased from ProteinTech (Rosemont, USA), triacsin c (10007448) was obtained from Cayman Chemical (MI, USA), and ABT-199 (S8048) was purchased from SELLECK (Shanghai, China).
2.10 TCF/LEF1 reporter assay
β-catenin activity was determined as luciferase transcription dependent on the T cell factor/lymphoid enhancer factor (TCF/LEF). The TCF/LEF1-luciferase reporter lentivirus was purchased from Genomeditech (Shanghai, China). Cells were infected with the reporter lentivirus and selected by 2 μg/mL puromycin for 2 days. Afterward, the cells were separately transduced with control and anti-ACSL5 shRNA lentivirus. At 72 h post transfection, the infection efficiency was estimated by determining the proportion of GFP-positive cells. Firefly luciferase activities were then assayed using the Dual-Luciferase Reporter Assay Kit (Promega, USA) according to protocol set by the manufacturer.
2.11 Metabolic labeling and click chemistry reaction
The cells were incubated with DMSO or 50 μmol/L palmitic acid analog ω-alkynyl palmitic acid Alk-C16 for 12 h. Labeled proteins bearing the alkyne group were then coupled with the azide group of biotin-azide or Azide Alexa Fluor 555 in click chemistry reaction. A click reaction solution was prepared in the following order: 0.1 mmol/L Azide Alexa Fluor 555 or biotin-azide, 1 mmol/L TCEP, 100 μmol/L TBTA, and 1 mmol/L CuSO4 in ddH2O.
Total palmitoylated proteins detected by flow cytometry: Alk-C16-labeled cells were fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.3% TritonX-100 for 10 min at room temperature. The cells were then treated with click reaction solution containing Azide Alexa Fluor 555 for 30 min in the dark at room temperature. The fluorescence intensity of palmitoylated proteins was analyzed using the flowJo software.
Total palmitoylated proteins by western blot: after incubation, the cells were lysed in a lysis buffer containing 50 mmol/L TEA-HCl, pH 7.4, 150 mmol/L NaCl, 1% TritonX-100, 0.1% SDS, 1 × HALT protease, and phosphatase inhibitor cocktail (#78442, Thermo Scientific, USA). Lysates were cleared by centrifugation (13 000× g for 15 min at 4 °C). Supernatants were collected, and protein concentrations were determined using the BCA assay. An aliquot of 200 μg of protein from each sample was adjusted to 50 μL using a lysis buffer and then incubated with a click reaction solution containing biotin-azide. Biotinylated proteins were precipitated using methanol/chloroform and washed with methanol prior to air drying. The protein pellets were dissolved in a 1× SDS sample loading buffer. The samples were then subjected to western blot and immunoblotted with streptavidin-HRP.
Palmitoylated Wnt3a detected by western blot: the labeled cells were lysed, and the extracts were divided into input and IP samples (at about 1:5 ratio). The IP samples were incubated with anti-Wnt3a antibody and protein A/G beads. The Wnt3a proteins in the immunocomplexes captured with beads were then reacted with biotin-azide, subjected to SDS-PAGE, and immunoblotted with streptavidin-HRP. The streptavidin-HRP was stripped for 10 min in a Restore PLUS Western Blot Stripping Buffer (Thermo Scientific, USA), washed, blocked, and then incubated with the anti-Wnt3a antibody (1:1000 dilution). The input samples were subjected to western blot and immunoblotted with an anti-Wnt3a and β-tubulin antibody.
2.12 Leukemia xenograft model
A total of 18 female NOD-PrkdcscidIL2rgtm1/Bcgen (B-NSG) (age, 6–8 weeks; weight, 18–22 g; Biocytogen, China) were randomly sorted into two groups. Negative control THP-1-luc cells or stable ACSL5-Knockdown (ACSL5-KD) THP-1-luc cells were transduced and purified by cell sorting with a GFP marker 3 days after infection. Afterward, an equal number (1 × 106) of sorted cells were injected into two groups of mice separately through their tail veins. For bioluminescence imaging, the mice were anaesthetized with 3% isoflurane and then administered 100 mg/kg D-luciferin (Promega, USA) by intraperitoneal injection. The tumor load was monitored by bioluminescence imaging at days 14, 21, 28, and 35 post-transplantation, and the survival of mice was also recorded. Three mice from each group were sacrificed at day 35 post-transplantation, and the others were fed continuously to measure their survival. Murine bone marrow cells were obtained from their femur and tibia bone marrow, and their spleen and liver were harvested and weighed. All experimental procedures were conducted in accordance with the institution guidelines for care and use.
2.13 Gene set enrichment analysis (GSEA)
RNA-seq data from the transcriptome profile in the AML dataset were downloaded from TCGA. GSEA was performed using GSEA 3.0 (BROAD Institute). Normalized gene expression profiles were ranked by a signal-to-noise metric, and the normalized enrichment scores (NES) were calculated with random gene set permutation 1000. Significance was considered at a nominal P value (Nom P value) of < 0.05 and a false discovery rate (FDR) of 0.25.
2.14 Statistical analysis
A base line data comparison was conducted based on Student’s t-test, Chi-square test, or one-way analysis of variance (ANOVA) depending on the needs. The survival data for the patients and mice were analyzed through the Cox regression model using the R package survival version 3.2-11. The experiments were repeated at least three times. The comparisons were performed using Student’s t-test or one-way ANOVA. The data were presented as mean ± SD or SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 indicated significant differences.
3 Results
3.1 High ACSL5 expression is associated with poor outcomes among AML patients.
To reveal the potential prognostic value of the ACSL5 mRNA expression for AML patients, we performed an explorative analysis of ACSL5 expression by exploring TCGA and GTEx datasets (via the GEPIA platform [
22]). The expression of ACSL5 increased in the AML samples compared with the healthy donors (
P < 0.05, Fig.1). The Kaplan–Meier survival analysis of the AML cohort from TCGA revealed that a high ACSL5 expression is associated with a poor overall survival for AML patients (Fig.1). The detailed clinical characteristics and multivariate analysis results of the TCGA cohort are shown in supplementary Tables S1 and S2, respectively. In addition, the CD34
+ hematopoietic stem cells from 6 healthy donors and the bone marrow mononuclear cells from 312 CN-AML patients were collected. The mRNA levels of ACSL5 in AML patients were significantly higher than those in healthy individuals as confirmed by RT-qPCR (Fig.1). We further explored the correlation between the ACSL5 level in AML bone marrow cells and the outcomes of the 312 CN-AML patients. The AML patients with high ACSL5 expression (
n = 156) had a relatively shorter overall survival (OS) (
P < 0.0001) and event free survival (EFS) (
P = 0.0004) compared with those in the low expression group (
n = 156) (Fig.1 and 1E). The baseline characteristics of the 312 CN-AML patients are provided in Tab.1. A multivariate analysis of the OS and EFS of our cohort showed that the ACSL5 expression remained a poor prognostic factor (OS: HR = 1.932 (1.253, 2.979); EFS: HR = 1.544 (1.040, 2.293)) and an independent predictor (OS:
P = 0.003; EFS:
P = 0.031) after adjusting for age, WBC, gene mutation (including
FLT3-
ITD mutation,
NPM1, DNMT3A, CEBPA, and
IDH1/2 mutations), and chemotherapy regimen (Tab.2).
3.2 ACSL5 deficiency suppresses cell growth in vitro
To further investigate the role of ACSL5 in AML, we examined the mRNA and protein expressions of ACSL5 in a panel of AML cell lines (Fig. S1A and S1B). Among these cell lines, THP-1 and OCI-AML2 with a relatively high ACSL5 level and KG-1 with a relatively low ACSL5 expression were chosen for genetic manipulation to dissect the function of ACSL5. The efficiency of knockdown or overexpression of ACSL5 was determined by RT-qPCR and Western blot analysis. The MTS and EdU assays revealed that the reduction in ACSL5 expression was accompanied by an instantly diminished proliferative capabilities of THP-1 and OCI-AML2 cells (Fig.2–2D). By contrast, the overexpression of ACSL5 in KG-1 cells significantly promoted cell proliferation compared with those transduced with an empty vector (Fig. S1C–S1E). Furthermore, Annexin V/PI staining followed by flow cytometric analysis showed increased cell apoptosis in ACSL5 knockdown cells (Fig.2 and 2F). The ACSL5-deficiency-induced apoptosis in THP-1 and OCI-AML2 cells was also evidenced by cleaved Caspase 3 and PARP (Fig.2). The knockdown of ACSL5 had no apparent effects on cell cycle progression (Fig.2 and 2I), and the ACSL5 overexpression did not produce an additional effect on cell apoptosis and cell cycle (Fig. S1F and S1G). Collectively, these results indicate that ACSL5 enhances AML cells expansion by promoting cell proliferation and inhibiting cell apoptosis.
3.3 Depletion of ACSL5 inhibits tumor growth in the AML xenograft model
To confirm the tumor-promoting role of ACSL5 in vivo, a xenograft mouse model was established by injecting THP-1-luc cells into B-NSG mice. The mice were randomly divided into two groups (nine mice per group) and were injected with control or ACSL5-knockdown THP-1-luc cells through their tail veins. Fluorescence imaging was performed at days 14, 21, 28, and 35 post-injection (Fig.3). The overall tumor burden of mice transplanted with ACSL5-knockdown cells was significantly lower than that observed in the control group (Fig.3). Moreover, the overall and median survival times of mice transplanted with ACSL5-KD cells were extended to 68 and 65 days, respectively, whereas that of the control group were merely 43 and 42 days (Fig.3). Three mice from each group were sacrificed at day 35 post-injection to evaluate the engraftment of human CD45 positive cells in BM cells and to weigh the livers and spleens. As expected, the proportion of human CD45 positive cells decreased in the BM of mice transplanted with ACSL5-KD cells compared with that in the BM of control mice (Fig.3). Likewise, the mice in the ACSL5-KD group indicated a significant reduction in their liver and spleen weights (Fig.3 and 3F).
3.4 ACSL5 modulates the activity of Wnt/β-catenin signaling by palmitoylation modification
To determine through which pathway ACSL5 promotes AML cell growth, we performed GSEA analysis based on mRNA expression data of an AML cohort from the TCGA database. Results of the GSEA analysis strongly implicated Wnt/β-catenin signaling as the main biological process affected by ACSL5 (Fig.4). Knockdown of ACSL5 significantly reduced the Wnt signaling response, which was accompanied by an increase in GSK-3β phosphorylation and a reduction in β-catenin accumulation in THP-1 and OCI-AML2 cells (Fig.4). We further used the TCF/LEF reporter assay to investigate whether ACSL5 influenced the activity of the Wnt/β-catenin signaling pathway, and the results revealed that the inhibition of ACSL5 significantly downregulated TCF/LEF-dependent luciferase activity (Fig.4). Meanwhile, the click chemistry assay revealed that the knockdown of ACSL5 induced a significant decrease in the palmitoylation of whole proteins (Fig.4) and the Wnt3a protein (Fig.4).
3.5 Triacsin c, the pan-ACS inhibitor, exerts a synergistic inhibition effect with ABT-199 in AML cells
To investigate the therapeutic potential of triacsin c (a pan-ACS inhibitor) in AML, we analyzed the responses of human AML cells to triacsin c in vitro. Triacsin c inhibited cell growth in six AML cell lines in a concentration-dependent manner at 48 h. Among the tested cell lines, the IC50 values of THP-1 and OCI-AML2 were 0.4671 ± 0.0531 μmol/L and 0.3683 ± 0.0232 μmol/L respectively, whereas those of the other AML cells are shown in supplementary Fig. S2A. Those cells treated with increasing concentrations of triacsin c showed an increased rate of apoptosis (Fig. S2B and S2C). We hypothesized a potential synergistic effect between triacsin c and ABT-199, the BCL2-selective inhibitor, that could induce a more intense cell apoptosis. To this end, we treated THP-1 and OCI-AML2 cells with triacsin c or ABT-199 either alone or in combination. Consistent with our hypothesis, we observed synergistic effects at triacsin c concentrations of 0.25 μmol/L to 1 μmol/L when combined with ABT-199 in both THP-1 and OCI-AML2 cells (Fig.5 and 5B). Moreover, the combined treatment significantly increased apoptosis as evidenced by the increase in Annexin V stained cells and cleaved Caspase 3 and PARP levels compared with cells treated with either triacsin c or ABT-199 (Fig.5, 5D and S5D). Moreover, triacsin c substantially inhibited cell growth in primary cells but did not exhibit cytotoxicity to normal CD34+ cells (Fig.5 and 5F). We further measured the sensitivity of triacsin c and the expression level of ACSL5 in AML primary cells (AML#6 to #15). However, we found no significant association between the ACSL5 level and IC50 of triacsin c in the Spearman rank correlation analysis (Fig. S4). In addition, the combination treatments of triacsin c or ABT-199 in primary cells (AML#16 to #18) were significantly more efficient than the single treatment (Fig.5‒5I).
4 Discussion
ACSL5 has a special position among acyl-CoA synthetases given its distinct subcellular localization on the mitochondrial membrane and strong preference for palmitate and palmitoleic acids [
4]. ACSL5 was identified as a functional regulator for cells to sustain their proliferation and resistance to cell death. Mashima
et al. reported that ACSL5 could help improve glioma cell survival under extracellular acidosis conditions [
8]. Moreover, long-chain acyl-CoA esters, as the products of ACSL5, are important intermediates in lipid biochemistry and have been reported to affect several aspects of cellular functions and signaling pathways [
23]. We demonstrated that ACSL5 facilitated the growth of AML cell lines
in vitro and
in vivo based on the stable ACSL5 knockdown or overexpression cells.
We also observed that the inhibition of ACSL5 increased the GSK-3β phosphorylation and degraded β-catenin, which in turn inhibited the activity of the Wnt signaling pathway. ACSL5 plays a crucial role in the palmitate metabolic process as it converts palmitate into its active form, namely, palmitoyl-CoA [
23]. Palmitoyl-CoA serves as a palmitate donor for Wnt3a palmitoylation. The knockout of Acsl5 showed a 60% reduction in jejunal palmitoyl-CoA synthesis rate and did not alter the mRNA expression of the other major isoforms of ACSL [
6]. Collectively, LC-MS/MS demonstrated a positive correlation between ACSL5 activity and acyl-CoA synthesis, particularly for less abundant sphingolipid species and saturated acyl-CoA species, such as palmitate [
24]. Therefore, we could reasonably assume that ACSL5 may play its function by directing the cellular metabolism of palmitate and regulating the palmitoylation level of key proteins.
Palmitoylation is a specific form of post-translational modification that involves the thioesterification of a 16-carbon saturated fatty acid (palmitate) [
25]. This palmitoylation can regulate the localization, accumulation, secretion, stability, and function of proteins by altering their membrane affinity [
26]. Indeed, the lipidation of some key signaling proteins, such as Hedgehog, Wnt, and RAS, is essential for their function both physiologically and in cancer [
25,
26]. The functional association of fatty acid metabolizing enzymes to the modification of Wnt proteins is of high importance in building a molecular link between cell intrinsic fatty acid metabolism and the critical signaling pathway. Our study demonstrated that the suppression of ACSL5 decreased the level of Wnt3a palmitoylation modification, which accounted for the inhibition of the Wnt/β-catenin signaling pathway.
Wnt signaling has been strongly associated with a variety of cancers [
27,
28]. The aberrant activation of Wnt signaling is a frequent yet heterogenous feature of AML [
29]. Wnt/β-catenin pathway is required in the self-renewal of leukemia stem cells [
30]. The canonical Wnt pathway is activated upon the binding of secreted Wnt proteins to its receptor Frizzled (FZD) in complex with co-receptors LPR5/6 at the plasma membrane. Subsequently, the β-catenin is dephosphorylated and stabilized from the destruction complex. β-catenin accumulates and translocates into the nucleus where it interacts with the TCF/LEF transcription factor family to drive a downstream gene expression [
29].
We found strong evidence to suggest that the palmitoylation in Wnt3a protein is involved in intracellular trafficking during the secretory process. Although Wnt3a was originally reported to be fatty acylated at Cysteine 77 [
31], Gao and Nile
et al. demonstrated that Cys 77 is unlikely to be available for fatty acylation [
32,
33]. Wnt3a was palmitoylated by O-acyl-transferase—porcupine at Serine 209 but not Cys77. The palmitoylation at Serine 209 of the Wnt3a protein was reported by metabolic labeling with a radiolabeled palmitate. The Wnt3a with acylation defect at Ser209 was not secreted from cells in culture and was instead retained in the endoplasmic reticulum. Ser209-dependent acylation may be required for targeting Wnt3a proteins to specific organelles or membrane components for secretion [
32,
34]. Therefore, we hypothesized that the altered palmitoylation of Wnt3a induced by ACSL5 knockdown was responsible for the defective Wnt/β-catenin response.
Accumulating evidence suggests that the dysregulation of ACSL5 is involved in various cancers, including glioma, colon cancer, and lung cancer [
8,
35–
37]. Therefore, ACSL5 was considered a diagnostic and prognostic marker and a potential therapeutic target against cancer. However, its clinical significance has not been elucidated in AML. In this study, we revealed for the first time that the expression of ACSL5 was elevated in BM samples of AML patients. A high expression of ACSL5 was also associated with poor overall survival in AML.
Triacsin c is an inhibitor of ACS family that competes with fatty acids for their catalytic domain [
38]. Triacsin c could induce massive apoptosis in glioma cells, whereas cell death was completely suppressed by the overexpression of ACSL5 [
39]. Furthermore, in terms of p53-defective tumor cell lines, triacsin c could selectively induce apoptosome-mediated cell death in tumor cells [
19,
39]. In line with these findings, the treatment of triacsin c induced cell apoptosis in AML cell lines and primary cells. ABT-199, an FDA-approved BCL-2 selective inhibitor, could disrupt the interaction between Bim and BCL-2, thereby freeing the former to induce apoptosis [
40]. Unlike most oncogenic proteins, BCL-2 dysregulation does not confer an advantage in cell growth or proliferation but instead allows cells that would normally undergo apoptosis to survive. Some cancer cells approach the threshold of apoptosis but are held back from death by BCL-2 family proteins that favor their survival [
41]. In our study, we found that the inhibition of ACSL5, either by triacsin c or shRNA lentivirus, significantly promotes cell apoptosis, hence leading us to wonder whether the addition of ABT-199, which directly targets apoptosis, could push cells on the verge of breakdown toward apoptosis. Triacsin c exhibited a synergistic effect with ABT-199, thereby suggesting that triacsin C has the potential to be an effective anti-leukemia agent alone or combined with ABT-199. In conclusion, we demonstrated for the first time that ACSL5 is a reliable prognosis marker and a potential therapeutic target for AML.
PDF
(5457KB)
