Inhibiting PLK1 induces autophagy of acute myeloid leukemia cells via mammalian target of rapamycin pathway dephosphorylation.

Decreased autophagy is accompanied by the development of a myeloproliferative state or acute myeloid leukemia (AML). AML cells are often sensitive to autophagy‑inducing stimuli, prompting the idea that targeting autophagy can be useful in AML cytotoxic therapy. AML NB4 cells overexpressing microtubule-associated protein 1 light chain 3-green fluorescent protein were screened with 69 inhibitors to analyze autophagy activity. AML cells were treated with the polo-like kinase 1 (PLK1) inhibitors RO3280 and BI2536 before autophagy analysis. Cleaved LC3 (LC3-II) and the phosphorylation of mammalian target of rapamycin (mTOR), adenosine monophosphate-activated protein kinase, and Unc-51-like kinase 1 during autophagy was detected with western blotting. Autophagosomes were detected using transmission electron microscopy. Several inhibitors had promising autophagy inducer effects: BI2536, MLN0905, SK1-I, SBE13 HCL and RO3280. Moreover, these inhibitors all targeted PLK1. Autophagy activity was increased in the NB4 cells treated with RO3280 and BI2536. Inhibition of PLK1 expression in NB4, K562 and HL-60 leukemia cells with RNA interference increased LC3-II and autophagy activity. The phosphorylation of mTOR was reduced significantly in NB4 cells treated with RO3280 and BI2536, and was also reduced significantly when PLK1 expression was downregulated in the NB4, K562 and HL-60 cells. We demonstrate that PLK1 inhibition induces AML cell autophagy and that it results in mTOR dephosphorylation. These results may provide new insights into the molecular mechanism of PLK1 in regulating autophagy.

Introduction

Autophagy, or autophagocytosis, is from the Greek auto-, ‘self’, and phagein, ‘to eat’. Autophagy is a conserved protein degradation pathway that plays important roles in mammalian cell survival, proliferation and differentiation, particularly within the hematopoietic system. Diminished autophagic flux results in the development of a myeloproliferative state or acute myeloid leukemia (AML) (1). In AML, there is copy number loss in autophagic genes such as BECN1 (2,3), ATG7 (1,4) and ATG5 (1,5). Decreased autophagy and the development of AML are related. BECN1 is a critical mediator that influences the onset and progress of autophagy, and there is a remarkable association between reduced BECN1 expression and FLT3-internal tandem duplication (ITD) mutation (6). ATG7 and ATG5 play important roles in autophagy and their loss of function in hematopoietic stem and progenitor cells (HSPCs) always leads to a lethal pre-leukemic phenotype in mice (1).

Recently, autophagy pathway inducers have shown promising effects for treating AML. Mammalian target of rapamycin (mTOR) signaling is a critical pathway in the biology of several cancers, including AML. Constitutive activation of the phosphatidylinositol 3-kinase (PI3K)/mTOR signaling pathway has been observed in different cancers and leukemias, including chronic myelogenous leukemia (CML), AML and acute lymphoblastic leukemia (ALL). The PI3K/mTOR pathway has always been considered a prosurvival factor in leukemia stem cells and leukemic precursors, and its inhibition has been regarded as an effective therapeutic approach (7). MLN0128 is a novel, recently developed mTOR kinase inhibitor that can disrupt survival signaling and triggers apoptosis in AML stem and AML progenitor cells (8). Abnormal mTOR activity contributes to chemotherapy resistance, and aberrant activation of the PI3K/mTOR pathway promotes sorafenib resistance in AML cells (9).

The serine/threonine protein kinase polo-like kinase 1 (PLK1), or serine/threonine-protein kinase 13 (STPK13), regulates multiple intracellular processes, including DNA replication, mitosis and stress response. PLK1 is expressed during mitosis and is overexpressed in multiple cancers, including breast cancer (10), prostate cancer (11), renal cancer (12) and neuroblastoma (13). PLK1 is also highly expressed in leukemia cell lines; PLK1 expression in patients with AML is significantly higher than in normal progenitors (14). Furthermore, PLK1 expression in normal or untransformed cells is much lower than in cancer cells, which renders PLK1 a suitable anticancer target (15,16). Downregulating or inhibiting the kinase activity of PLK1 induces cell cycle arrest and apoptosis in most cancer cell types in vitro and in vivo (17–20). The potential of PLK1 inhibitors as cancer therapeutics has been investigated widely. The PLK1 inhibitor volasertib has shown considerable promise in clinical studies of AML, having reached phase III trials (21,22). Other PLK1 inhibitors, including GSK461364A, TKM-080301, GW843682, purpurogallin and poloxin are in early clinical development (23).

To date, the molecular function of PLK1 in AML cell autophagy is unclear. In our study, the autophagy-related effect of PLK1 was evaluated in AML cells to characterize its preclinical efficacy.

Materials and methods

Animal and human rights statement

The studies have been performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments. Ethical approval was provided by the Children's Hospital of Soochow University Ethics Committee (nos. SUEC2008-011 and SUEC2000-021).

Cell and culture conditions

Leukemia cell lines HL-60 and K562 were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). NB4 cell line (gifts from Hematology Institute of Soochow University). All cell lines were maintained at 37°C in the RPMI-1640 (Gibco Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen Life Technologies, Carlsbad, CA, USA). Sixty-nine inhibitors ABT-263, ABT-737, YM155, SK1-I, SKI-5C, 17-AAG, XAV-939, AC220, tosedostat (CHR2797), VER-50589, FH535, G-749, BV-6 (apoptosis and anti-apoptosis); rapamycin, valproic acid, 3-methyladenine (3-MA), BEZ235, HS-173, pilaralisib (autophagy); SP600125, elesclomol, BAY 11–7082, ipatasertib, SB202190, PD98059, LY294002, INCB018424, SH-4-54, AT13148, JNK inhibitor IX, PX-478 2HCl (oxidative stress and MAPK pathway); BI 2536, PF-3758309, nutlin-3, MI-773, YH239-EE, XL-413, MLN0905, SBE13 HCL, RO3280, volasertib, nutlin-3b (cell cycle); JIB-04, GSK J1, GSK J4, GSK 126, LBH589, SGC-CBP30, 4SC-202 (histone modification); KPT-276, KPT-330, KPT-185, KPT-335 (CRM1); CW069, TAPI-1, INH6, ODM-201, ESI-09, EW-7197, GDC-0623, AZD6738, LY3009120, SB-3CT, INH1, XMD8-92, LY2584702, ML323 (other targets). Tosedostat (CHR2797) was purchased from Molbase Chemicals (Shanghai, China). Other inhibitors were purchased all from Selleck Chemicals (West Paterson, NJ, USA) and was dissolved in dimethyl sulfoxide (DMSO) (cat. no. D4540; Sigma-Aldrich, St. Louis, MO, USA).

Expression of hLC3 in leukemia cells with UBC-LC3-GFP lentivirus

UBC-LC3-GFP lentivirus was purchased from Shanghai Genechem Co., Ltd. (Shanghai, China) (24). Lentivirus infection was according to the manufacturer (Shanghai GeneChem Co, Ltd.) at a final concentration of 100–200 multiplicity of infection (MOI). Aggregation of LC3 was observed when the autophagy occurred in cells. LC3-GFP fusion protein is diffused in the cytoplasm when the cells occur autophagy. LC3-GFP fusion protein translocates to autophagic membranes, forming a bright green fluorescence spots under the fluorescence confocal microscopy (Olympus; Olympus Corporation, Tokyo, Japan) when the cells are autophagic, a spot is equivalent to an autophagosome thus can be counted to evaluate the level of autophagy activity.

Transmission electron microscopy (TEM)

The presence of autophagosomes in TEM is the standard for detecting autophagy. TEM analysis protocol was according to a previous report (25). Briefly, leukemia cells grown on 6-well plates were treated with RO3280, BI2536 or the same volume of DMSO and 4 h later were harvested and washed with phosphate-buffered saline (PBS) prior to fixing in fixative buffer. Subsequently, cells were collected and suspended with 2.5% glutaraldehyde. Then treated with 2% osmium tetroxide in 0.1 M sodium cacodylate buffer, and embedded in resin. Finally, the samples were sliced and processed for TEM analysis (Hitachi electron microscope H-600; Hitachi, Ltd., Tokyo, Japan).

Interfering expression of PLK1 in leukemia cells with RNAi lentivirus

RNAi lentivirus for PLK1 was purchased from Shanghai GeneChem Co, Ltd. RNAi products consist of target-specific lentivirus designed to knock down PLK1 expres- sion sequences are: 5′-CCGAGTTATTCATCGAGAC-3′. The control sequences are: 5′-TTCTCCGAACGTGTCACGT-3′. Lentivirus infection was according to the manufacturer (Shanghai Genechem Co., Ltd.) at a final concentration of 100–200 MOI. PLK1 interference efficiency was measured by western blotting at 3 days after transfection. The rest of the cells were harvested for further analysis.

Cell proliferation analysis

Leukemia cells (2×104) were seeded in 96-well plates overnight and incubated with DMSO, or increasing concentrations of RO3280, BI2536 24 to 96 h. Each drug concentration was performed replicated 2 times. Then, 10 µl Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc., Kumamoto, Japan) solution was added to each well, incubated at 37°C for 2 to 4 h and the optical density (OD) values were measured at 450 nm using a scanning multi-well spectrophotometer (Bio-Rad Model 550; Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Apoptosis assay

Apoptosis assay was according to the manual operation of BD Annexin V staining kit (cat. no. 556420; BD Biosciences, Franklin Lakes, NJ USA). Briefly, cells were washed 2 times with cold PBS and then were resuspend in 1X binding buffer at a concentration of ~1×106 cells/ml, and 100 µl of the solution (~1×105 cells) was transfered to a 5-ml culture tube. Annexin V and PI (5 µl/test) was added. Cells were gently mixed and incubated for 15 min at room temperature in the dark. Four hundred microliters of 1X binding buffer was added to each tube. Cells were analyzed by flow cytometry within 1 h (26).

Western blotting

For western blotting, protocol was introduced as described (26). Cellular proteins were blocked and then probed with antibodies against PARP (cat. no. 9542S, 1:1,000), PLK1 (cat. no. 4535S, 1:1,000), beclin-1 (cat. no. 3495S, 1:1,000), SQSTM1/p62 (cat. no. 8025S, 1:1,000), LC3A/B (cat. no. 4108S, 1:1,000), mTOR (cat. no. 2983S, 1:1,000), phospho-mTOR (Ser2448) (cat. no. 5536S, 1:1,000), adenosine monophosphate-activated protein kinase (AMPK)α (cat. no. 5831S, 1:1,000), phospho-AMPKα1 (Ser485) (cat. no. 2537S, 1:1,000), phospho-AMPKα1 (Thr172) (cat. no. 2537S, 1:1,000), Unc-51-like kinase 1 (ULK1) (cat. no. 6439S, 1:1,000), phospho-ULK1 (Ser317) (cat. no. 12753S, 1:1,000), phospho-ULK1 (Ser757) (cat. no. 14202S, 1:1,000), phospho-ULK1 (Ser555) (cat. no. 5869S, 1:1,000) all these antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA) and GAPDH (1:5,000; Sigma-Aldrich). Relative expression of p-mTOR/mTOR was analyzed with ImageJ software according to the user guide.

Patients and samples

Bone marrow specimens were obtained at the time of diagnosis during routine clinical assessment of 140 pediatric patients with AML, who presented at the Department of Hematology and Oncology, Children's Hospital of Soochow University between 2008 and 2013. Ethical approval was provided by the Children's Hospital of Soochow University Ethics Committee (nos. SUEC2008-011 and SUEC2000-021), and informed consent was obtained from the parents or guardians. AML diagnosis was made in accordance with the revised French-American-British (FAB) classification. The main clinical and laboratory features of the patient cohort are summarized in Table I.

Table I.

Association of PLK1 expression with clinico-pathological characteristics in 140 pediatric AML samples.

		PLK1 expression (n)
Clinical variables	No. of patients	Low	High	P-value
Gender				0.888
Male	56	34	22
Female	84	50	34
Age (years)				0.334
<6	73	41	32
≥6	67	43	24
Leukocyte (/µl)				0.673
>10,000	83	51	32
≤10,000	57	33	24
FAB				0.027
M1-M6	123	78	45
M7	17	6	11
Cytogenetics				0.382
Favorable	70	43	27
Intermediate	38	25	13
Unfavorable	32	16	16
MRD				0.038
<0.25%	65	45	20
≥0.25%	75	39	36

PLK1, polo-like kinase 1; AML, acute myeloid leukemia; FAB, French-American-British; MRD, minimal residual disease.

Quantitative reverse transcription-polymerase chain reaction (PCR) for PLK1

Quantitative real-time PCR was performed to determine the expression levels of PLK1 genes. Total RNA was reverse transcribed using the Reverse Transcription kit, according to the manufacturer's instructions (Applied Biosystems, Foster City, CA, USA). The real-time PCR primers used to quantify GAPDH expression were forward, 5′-AGAAGGCTGGGGCTCATTTG-3′ and reverse, 5′-AGGGGCCATCCACAGTCTTC-3′; and for PLK1 were forward, 5′-CTCAACACGCCTCATCCTC-3′ and reverse, 5′-GTGCTCGCTCATGTAATTGC-3′. Expression of PLK1 was normalized to endogenous GAPDH expression.

Statistical analysis

Each experimental condition was performed three times, and these replicates are presented in the results. All values are presented as means ± SEM. Student's paired t-test was applied to reveal statistical significance. P-values <0.05 were considered significant. Statistical analyses were performed using SPSS software for Windows (version 11.5; SPSS, Inc., Chicago, IL, USA).

Results

Screening of novel autophagy inducers of AML cells

As autophagy pathway inducers have shown promising effects for treating AML, NB4 AML cells overexpressing LC3-GFP fusion protein were screened with 69 inhibitors (1 µM each; Selleck Chemicals, Houston, TX, USA) for 4 to 12 h before analysis. LC3 aggregation in the cells was observed during autophagy. LC3-GFP translocation to the autophagic membranes formed bright green fluorescent spots under fluorescence confocal microscopy when the cells underwent autophagy; one spot was equivalent to an autophagosome, which were counted to evaluate the level of autophagy activity (Fig. 1A). Autophagy activity was calculated as autophagy spots per number of cells. Our screening experiment showed that several inhibitors had promising autophagy inducer effects. Autophagy activity >5 in the inhibitors BI2536, MLN0905, SK1-I, SBE13 HCL and RO3280 (Fig. 1B-D). Rapamycin was used as the positive control, where its autophagy activity was ~5. As reported previously, SK1-I is a novel inhibitor of sphingosine kinase 1 (SPHK1) (27), an evolutionarily conserved lipid kinase responsible for converting sphingosine to sphingosine-1-phosphate (S1P). SPHK1 is overexpressed in many cancers and therefore serves as a cancer prognostic biomarker. BI2536, MLN0905, SBE13 HCL and RO3280 are all PLK1 inhibitors. PLK1 is always highly expressed in leukemic cell lines and is overexpressed in a majority of samples from patients with AML compared with normal progenitors (14). Yet, its molecular function in the autophagy of AML cells is still unclear.

Figure 1.

Novel autophagy inducer screen in AML cells. (A) Cells treated with rapamycin (positive control), DMSO (negative control) and 69 inhibitors observed under fluorescence confocal microscopy to analyze autophagy activity. White arrows indicate autophagosomes. (B-D) Autophagy activity was calculated as autophagy spots per number of cells. Our screening experiment showed that several inhibitors had promising autophagy inducer effects. Five-fold change in the figure is marked with dotted line. The autophagy activity of SK1-I, BI2536, MLN0905, SK1-I, SBE13 HCL and RO3280 was >5. AML, acute myeloid leukemia; DMSO, dimethyl sulfoxide.

PLK1 inhibitors induce AML cell autophagy

To confirm the induction of autophagy by the PLK1 inhibitors, we treated NB4 cells with 200 and 500 nM RO3280 or BI2536 for 4 h before analysis (Fig. 2A), and found that autophagy activity levels were increased (500 nM BI2536: 9.73±1.89 vs. DMSO: 0.73±0.45, p=0.011; 500 nM RO3280: 11.63±2.48 vs. DMSO: 0.27±0.15, p=0.015). LC3 is converted from the inactive form (LC3-I) to the cleaved form (LC3-II) during autophagy, rendering LC3 a specific marker for detecting autophagy. Western blotting showed more LC3-II in NB4 cells treated with RO3280 and BI2536. More LC3-II was observed in the 100–500 nM RO3280 group and in the 200–500 nM BI2536 group (Fig. 2B). Beclin-1 upregulation and SQSTM1/p62 downregulation was also observed in the PLK1 inhibitor treatment groups. TEM was used to examine the autophagosomes in RO3280-treated cells (Fig. 2C). Compared with the DMSO control group, there were more autophagosomes in RO3280-treated NB4 cells. These results are consistent with the autophagy activity analysis involving LC3-GFP. Detecting autophagy is related to correct timing. We treated NB4 cells with 0.5 µM RO3280 or BI2536 continuously 4 to 12 h, and observed their induction of autophagy at different time-points (Fig. 3A). There was more LC3-II in the NB4 cells treated with 0.5 µM RO3280 or BI2536. These results suggest that PLK1 inhibitors induce significant AML cell autophagy.

Figure 2.

PLK1 inhibitors induce AML cell autophagy. (A) Fluorescence confocal microscopy images of autophagy activity in NB4 cells treated with 200 and 500 nM RO3280 or BI2536 for 4 h. White arrows indicate autophagosomes. Analysis of autophagy activity in NB4 cells treated with RO3280 or BI2536. **p<0.01. (B) Western blotting showing more LC3-II in cells treated with RO3280 and BI2536. Beclin-1 was upregulated and SQSTM1/p62 was downregulated in the PLK1 inhibitor group. (C) TEM examination of autophagosomes in cells treated with 500 nM RO3280 for 4 h. Compared with the control cells (left, magnified inset), autophagosomes (red arrows) were rapidly increased in the RO3280-treated NB4 cells (right, magnified inset). PLK1, polo-like kinase 1; AML, acute myeloid leukemia; TEM, transmission electron microscopy.

Figure 3.

PLK1 inhibition by RNA interference induces more LC3-II in AML cells. (A) NB4 cells were treated with 500 nM RO3280 or BI2536 for 4 to 12 h. There was more LC3-II in NB4 cells treated with 500 nM RO3280 or BI2536. (B) Western blotting of PLK1 and LC3-I/LC3-II in NB4, K562 and HL-60 cells transfected with RNA interference lentivirus. (C) NB4, K562 and HL-60 cells transfected with RNA interference lentivirus observed under fluorescence confocal microscopy for analysis of autophagy activity. White arrows indicate autophagosomes. Autophagy activity in NB4, K562 and HL-60 cells. **p<0.01. PLK1, polo-like kinase 1; AML, acute myeloid leukemia.

PLK1 inhibition by RNA interference induces AML cell autophagy

Inhibitors tend to have several candidate targets, and their molecular function is influenced by other candidate targets, especially homologous molecules. The molecular function of PLK1 inhibitors is influenced by the candidate targets PLK2, 3 and 4. To confirm the effect of PLK1, we inhibited its expression with RNA interference. When PLK1 expression was downregulated, western blotting detected more LC3-II (Fig. 3B). LC3-GFP analysis showed increased autophagy activity levels when PLK1 expression was significantly downregulated in the three leukemia cell lines (Fig. 3C). The autophagy activity was calculated and results were as follows: NB4 cells, Si-PLK1: 11.23±2.43 vs. Si-Nc: 0.33±0.15 (p=0.015); K562 cells, Si-PLK1: 9.73±1.89 vs. Si-Nc: 0.63±0.49 (p=0.011); HL-60 cells, Si-PLK1: 6.73±0.90 vs. Si-Nc: 0.40±0.10 (p=0.002), showing that PLK1 inhibition by RNA interference significantly induces AML cell autophagy (Fig. 3C).

PLK1 inhibition induces AML cell autophagy via mTOR dephosphorylation

We demonstrated that PLK1 inhibition induces significant leukemia cell autophagy. We analyzed the phosphorylation of these three kinases in NB4 cells treated with RO3280 and BI2536, and found that mTOR phosphorylation was reduced significantly in NB4 cells treated with RO3280 and BI2536 (Fig 4A). The dephosphorylation of mTOR is a crucial step in the induction of autophagy in eukaryotes. ULK1 phosphorylation at Ser317 was also reduced significantly in NB4 cells treated with RO3280 and BI2536. AMPK levels and AMPK phosphorylation levels did not exhibit significant changes. To confirm the effect of PLK1, RNA interference lentivirus transfection of NB4, K562 and HL-60 cells was used to significantly downregulate PLK1 expression, which was followed by significantly reduced mTOR phosphorylation (Fig. 4B). These results imply that PLK1 inhibition induces AML cell autophagy via mTOR dephosphorylation.

Figure 4.

PLK1 inhibition induces AML cell autophagy via mTOR dephosphorylation. (A) Western blotting of mTOR, AMPK and ULK1 expression levels and phosphorylation in NB4 cells treated with RO3280 or BI2536. There was significantly decreased mTOR phosphorylation and ULK1 phosphorylation at Ser317. (B) Western blotting of mTOR expression levels and phosphorylation in NB4, K562 and HL-60 cells transfected with PLK1 RNA interference lentivirus at 100 and 200 MOI. Relative expression of p-mTOR/mTOR was analyzed with ImageJ software according to the user guide. **p<0.01. PLK1, polo-like kinase 1; AML, acute myeloid leukemia; mTOR, mammalian target of rapamycin; AMPK, adenosine monophosphate-activated protein kinase; ULK1, Unc-51-like kinase 1.

PLK1 is an excellent anticancer target in pediatric AML

PLK1 inhibition induces significant apoptosis in several cancer cell types. The relationship between autophagy and apoptosis following PLK1 inhibition in AML cells remains unclear. Our analysis showed that for 4 and 8 h, NB4 cell apoptosis was almost identical between the DMSO group and 200 nM RO3280 group, suggesting that autophagy occurs earlier than apoptosis when cells are treated with PLK1 inhibitors. Apoptosis increased significantly when the NB4 cells were treated with RO3280 for 12 h, and it was inhibited by 5 mM 3-MA, an autophagy inhibitor (Fig. 5A). Autophagy and apoptosis induced by PLK1 inhibition are closely related, and the specific molecular mechanism warrants further study. Proliferation analysis showed that PLK1 inhibition significantly inhibited NB4 cell proliferation (Fig. 5B and C).

Figure 5.

PLK1 is an excellent anticancer target for pediatric AML. (A) Analysis of NB4 cell apoptosis treated with RO3280 or 3-MA. At 4 and 8 h, NB4 cell apoptosis was almost identical between the DMSO and 200 nM RO3280 group. NB4 cell apoptosis increased significantly following 12-h treatment with RO3280, and was inhibited significantly by 5 mM 3-MA. **p<0.01. (B) Analysis of NB4 cell proliferation following treatment with 200 nM RO3280. (C) Analysis of NB4 cell proliferation following transfection with PLK1 RNA interference lentivirus at 100 MOI. *p<0.05, **p<0.01. (D) The prognostic significance of PLK1 expression was assessed in 140 Chinese patients with pediatric AML with clinical follow-up records. Kaplan-Meier survival analysis revealed shorter survival in the patients with high PLK1 expression in tumors (p=0.007). (E) ROC curve analysis showing that the area under the curve of cytogenetics, MRD, and PLK1 is 0.695, 0.763 and 0.613, respectively. PLK1, polo-like kinase 1; AML, acute myeloid leukemia; 3-MA, 3-methyladenine; DMSO, dimethyl sulfoxide; MOI, multiplicity of infection; ROC, receiver operating characteristic; MRD, minimal residual disease.

Patients were divided into two groups with average value of PLK1 expressions (10.32×10−5). Results revealed that PLK1 expression was related with FAB subtype and minimal residual disease (MRD; Table I). Kaplan-Meier survival analysis revealed shorter survival in patients with high PLK1 expression in tumors (p=0.007; Table II and Fig. 5D). Multivariate analysis revealed that PLK1 expression is a near independent prognostic factor in pediatric AML patients (p=0.066; Table III). Receiver operating characteristic (ROC) curve analysis showed that the area under the curve for cytogenetic, MRD, and PLK1 expression was 0.695, 0.763 and 0.613, respectively (Fig. 5E). In conclusion, our results demonstrate that PLK1 is an excellent anticancer target in pediatric AML.

Table II.

Association of PLK1 expression with Kaplan-Meier survival in 140 pediatric AML samples.

Variables	No. of patients	Over survival (mean ± SE)	P-value
Cytogenetics			<0.001
Favorable	70	48.277±2.857
Intermediate	38	34.785±2.840
Unfavorable	32	13.477±2.245
FAB			0.001
M1-M6	123	40.150±2.344
M7	17	14.120±3.343
Leukocyte (µl)			0.641
>10,000	83	34.611±2.472
≤10,000	57	36.893±3.431
MRD			<0.001
<0.25%	65	53.085±2.622
≥0.25%	75	25.061±2.358
PLK1 expression			0.007
Low <10.32×10−5	84	42.253±2.778
High ≥10.32×10−5	56	30.214±3.521

PLK1, polo-like kinase 1; AML, acute myeloid leukemia; FAB, French-American-British; MRD, minimal residual disease.

Table III.

Cox multivariate analysis of PLK1 expression and clinico-pathological features in pediatric AML.

Variables	Odds ratio	EXP(B) 95% CI	P-value
Cytogenetics
Favor. vs. intermed. and unfavorable	7.248	2.253 (1.247–4.070)	0.007
MRD
<0.25 vs. ≥0.25%	11.208	3.346 (1.650–6.786)	0.001
Leukocyte (µl)
>10,000 vs. ≤10,000	0.252	1.141 (0.681–1.912)	0.615
FAB classification
M7 vs. M1-M6	6.102	2.233 (1.180–4.222)	0.014
PLK1 expression
Low vs. high	3.370	1.605 (0.968–2.661)	0.066

PLK1, polo-like kinase 1; AML, acute myeloid leukemia; CI, confidence interval; FAB, French-American-British; MRD, minimal residual disease.

Discussion

Autophagy functions as a means of managing cellular stresses and plays a substantial role in tumor cell survival. Decreased autophagic flux results in the development of a myeloproliferative state and AML. In AML, there is copy number loss in the autophagic genes BECN1, ATG7 and ATG5. AML cells often remain sensitive to autophagy-inducing stimuli, leading to the idea that harnessing autophagy can be useful in AML cytotoxic therapy. NB4 AML cells overexpressing LC3-GFP were treated with 69 inhibitors for 4 to 12 h before analysis, and several had promising autophagy inducer effects, where the autophagy activity of BI2536, MLN0905, SK1-I, SBE13 HCL and RO3280 was >5. Very interestingly, these inhibitors all had the same target, PLK1.

PLK1 is expressed during mitosis and is overexpressed in multiple cancers, including acute leukemia (28). PLK1 is also overexpressed in a majority of samples from patients with AML compared with normal progenitors. However, the prognostic value of PLK1 in pediatric AML is still unclear to date. PLK1 supports centrosome functional maturation in late G2 prophase and the establishment of the bipolar spindle. These checkpoints, which occur at G2/M transition, are tightly regulated by nuclear serine/threonine kinases, including cyclin-dependent kinases (CDKs), PLKs and Aurora kinases. Evidence suggests that PLK1 can inhibit the transactivation and proapoptotic functions of p53 by physical interaction and phosphorylation (29). Overexpression studies of PLK1 in the NIH3T3 cell line showed that these cells become capable of forming foci and growing in soft agar, and more importantly, can form tumors in nude mice due to PLK1 overexpression (30). In our study, Kaplan-Meier survival analysis revealed shorter survival in patients with high PLK1 expression in tumors. PLK1 inhibition induced significant AML cell apoptosis. These results suggest that PLK1 is an excellent anticancer target in pediatric AML. In this study, we demonstrate that PLK1 inhibition induces AML cell autophagy. Autophagy activity levels were increased when NB4 cells were treated with the PLK1 inhibitors RO3280 and BI2536. Inhibiting PLK1 expression in NB4, K562, and HL-60 cells with RNA interference led to more LC3-II, and autophagy activity was increased when PLK1 expression was significantly downregulated in these cells. Other members of PLK1 family have been reported with a relationship with autophagy. Overexpression of PLK3 mediates the degradation of abnormal prion proteins dependent on chaperone-mediated autophagy (31). PLK2 has a relationship with α-synuclein which plays an important role in autophagy (32). These results indicate that PLK1 family members may all have an important role in autophay. Also, this is the first study of autophagy induction in AML cells following PLK1 inhibition.

To date, the molecular mechanism of PLK1 inhibition in the induction of AML cell autophagy remains unknown. mTOR, or FK506-binding protein 12 (FKBP12)-rapamycin- associated protein 1, is a serine/threonine protein kinase from the family of PI3K-related kinases. In mammals, amino acid sensing and other signals such as reactive oxygen species (ROS) and growth factors regulate the activity of the protein kinases mTOR and AMPK, which regulate autophagy through inhibitory phosphorylation of ULK1 and 2. Induction of autophagy results in ULK kinase dephosphorylation and activation. The active ULK relocalizes to the site of autophagosome initiation, contributing to activation of the downstream autophagy components. Accordingly, mTOR, AMPK and ULK1 are important regulators of autophagy initiation. As reported previously, under nutrient sufficiency, high mTOR activity prevents ULK1 activation by phosphorylating and disrupting the interaction between ULK1 and AMPK (33–35). This coordinated phosphorylation is important for ULK1 in autophagy induction. We analyzed the protein phosphorylation of mTOR, AMPK and ULK1, three important regulators of autophagy, and found that mTOR phosphorylation was reduced significantly in NB4 cells treated with RO3280 and BI2536. To confirm the effect of PLK1 in NB4, K562 and HL-60 cells, mTOR phosphorylation was also reduced significantly when PLK1 expression was downregulated using RNA interference. mTOR, a serine/threonine protein kinase with a molecular size of ~300 kDa, belongs to the phosphatidylinositol 3-kinase-related kinase (PIKK) family. Its activity is inhibited under nutrient starvation and other stimuli of autophagy induction, and its dephosphorylation is a crucial step in the induction of autophagy in eukaryotes. The PI3K/AKT/mTOR pathway is very important in cell growth, survival, apoptosis, angiogenesis and phagocytosis. Disorder of this pathway results in disease, including cancer, neurological disease and autoimmune disease. The PI3K/AKT/mTOR pathway is constitutively activated in ~50–80% of AML cases (36), and its activation is associated with poor prognosis of AML. PI3K/AKT/mTOR may be an important therapeutic target in AML. Unfortunately, targeting the mTOR pathway with single-agent rapamycin in clinical trials was unsuccessful (37). As an alternative to adding chemotherapy, it is useful to target both the mTOR pathway and other cancer targets to generate more effective responses (38,39). As PLK1 inhibitors inhibit the activity of PLK1 and mTOR, two important pathways in AML, the present findings imply that PLK1 inhibitors may have good clinical application in AML. The details of PLK1 regulation of mTOR phosphorylation are still unknown and warrant further studies.

In conclusion, we demonstrate that PLK1 is an excellent anticancer target in pediatric AML; its inhibition induces autophagy of AML cells and results in mTOR dephosphorylation in AML cells. These results may provide new insights into the molecular mechanism of PLK1 in autophagy regulation.