Low-dose arsenic-mediated metabolic shift is associated with activation of Polo-like kinase 1 (Plk1)

Abstract

Arsenic is a well-established human carcinogen associated with cancers of the skin, liver, lung, kidney, and bladder. Although numerous carcinogenic pathways have been proposed, the molecular mechanisms underlying arsenic-associated cancer etiology are still elusive. The cellular responses to arsenic exposure are dose dependent. It was recently shown that low-dose arsenic leads to a metabolic shift from mitochondrial respiration to aerobic glycolysis via inactivation of tumor suppressor p53 and activation of NF-κB. However, how inactivation of p53, activation of NF-κB, and metabolic change are coordinated in response to low-dose arsenic exposure is still not completely understood. Polo-like kinase 1 (Plk1) is a well- documented regulator in many cell cycle-related events. Herein, we showed that low-dose arsenic leads to elevation of Plk1 in an NF-κB-dependent manner and that elevation of Plk1 contributes to the metabolic change from oxidative phosphorylation to glycolysis via activation of the PI3K/AKT/mTOR pathway. Furthermore, we showed that inhibition/depletion of Plk1 reverses low-dose arsenic-associated phenotypes, including enhanced cell proliferation, activation of the PI3K/AKT/mTOR pathway, and increased glycolysis. Finally, inhibition of the PI3K/AKT/mTOR pathway also antagonizes the enhanced glycolytic influx due to low-dose arsenic exposure. Our studies support the notion that Plk1 likely plays a critical role in cellular responses to low-dose arsenic.

Abbreviations

Dox

doxorubicin

ECAR

extracellular acidification rate

GTT

glucose tolerance test

mTOR1

mammalian target of rapamycin complex I

MEF

mouse embryo fibroblast

PDTC

pyrrolidine dithiocarbamate

PI3K

phosphatidylinositol 3-kinase

Plk1

polo-like kinase 1

PTEN

phosphatase and tensin homolog; Plk1-iKD, Plk1-inducible knockdown

2-DG

2-deoxy-glucose.

Introduction

Chronic exposure to arsenic, which occurs worldwide primarily through consumption of contaminated drinking water from natural geological sources, causes an increase in the incidence of cancers in various organs, such as skin, kidney, liver, and bladder.^1-4 On applying the World Health Organization (WHO) provisional guideline for drinking water of 10 μg/L (∼0.5 μM) of arsenic, a worldwide population of more than 100 million people are at risk, and more than 2.5 million people use water containing arsenic higher than 25 μg/L (∼1.25 μM). Therefore, it is urgent to understand molecular mechanism of arsenic-associated cancer etiology. Dependent on the dosage, cellular responses to arsenic exposure can be quite different. For example, high-dose arsenic (5 μM for 24 h) leads to mitotic arrest, followed by cell death.⁵ Low-dose arsenic (50 nM for 12 h) actually induces chemotherapy protection, since it causes cytoplasmic accumulation of tumor suppressor p53, thus antagonizing DNA damage-induced p53 response in normal tissues during chemotherapy.^6-8 It was further shown that low-dose arsenic-induced chemotherapy protection is due to p53/NF-κB-mediated metabolic switch from mitochondrial respiration to aerobic glycolysis.⁶ Despite these progresses, the detailed mechanisms underlying low-dose arsenic-associated phenotypes, such as p53 inactivation, NF-κB activation, and metabolic switch, are still not completely understood. The data presented here aims to address this critical issue.

Arsenic can alter the phosphorylation status of signaling molecules by activating specific protein kinases. For example, arsenic exposure leads to AKT-mediated phosphorylation of filamin A, resulting in enhanced cells migration.⁹ It was recently shown that HSP70 is a substrate of polo-like kinase 1 (Plk1) in high-dose (5 μM for 24 h) arsenic trioxide-induced mitotically arrested cell and that its phosphorylation contributes to attenuation of arsenic-induced mitotic abnormalities.¹⁰ Whether Plk1 also plays a critical role in low-dose (lower than 1 μM) arsenic-mediated phenotypes is not known. Plk1 has well documented roles in many mitotic-related events, such as centrosome maturation, bipolar spindle formation, sister chromatid segregation and cytokinesis.¹¹ Overexpression of Plk1 has been found in many cancer cell lines and neoplastic tissues, thus it is generally believed that Plk1 elevation is oncogenic. However, increasing evidence suggests that Plk1 might have many functions besides mitosis.¹² For example, it was recently reported that Plk1 phosphorylation of PTEN leads to its inactivation, activation of PI3K/AKT/mTOR pathway, metabolic switch from mitochondrial respiration to glycolysis.¹³ Herein, we report that Plk1 is a critical downstream target of activated NF-κB caused by low-dose arsenic and that Plk1-dependent activation of the PI3K/AKT/mTOR pathway contributes to the metabolic switch from oxidative phosphorylation to aerobic glycolysis.

Results

Low-dose arsenic induces a metabolic shift in mice

It was recently reported that low-dose arsenic treatment results in increased glycolysis.^6,8 However, the molecular mechanism underlying this intriguing observation is still elusive. To systematically study the low-dose arsenic-associated metabolic shift, we first compared glucose homeostasis of mice fed with 0.01% sodium arsenite-containing water for 16 weeks with age-matched mice that were fed with regular water by performing glucose tolerance test (GTT). An intraperitoneal glucose challenge revealed an improved glucose tolerance in mice pre-treated with sodium arsenite, as arsenic-treated mice showed lower glucose levels (faster clearance) than control mice after injection (Fig. 1A). Next, indirect calorimetry analysis revealed that arsenic-treated mice showed higher energy expenditure than control mice, supported by both increased O₂ consumption (Fig. 1B) and CO₂ production (Fig. 1C). Third, to confirm that low-dose arsenic is indeed involved in energy metabolism, we derived MEF cells from mice and treated MEFs with 0.3 μM of sodium arsenite. As expected, arsenic-treated MEFs exhibited both increased glucose consumption (Fig. 1D) and lactate production (Fig. 1E).

Figure 1.

Low-dose arsenic induces a metabolic shift. (A) Low-dose arsenic treatment leads to increased glucose uptake. Mice were treated with 0.01% (24 μM) sodium arsenite for 16 weeks, starved for 16 h, and injected with glucose (2 mg per gram of body weight), followed by measurement of blood glucose levels at different post-injection times. (B and C) Indirect calorimetry of mice treated with low-dose arsenic. The mice were prepared as in (A) and subjected to indirect calorimetry for 48 hours in metabolic chambers. Oxygen consumption (B) and carbon dioxide release (C) were determined in control and arsenic-treated mice. (D and E) Low-dose arsenic leads to increased glycolysis. After MEFs (mouse embryo fibroblasts) were treated with sodium arsenite at 0.3 μM for 24 h, the media were collected for measurement of glucose consumption (D) and lactate production (E). (F) Plk1 is elevated upon low-dose arsenic treatment in vivo. Lysates prepared from the indicated tissues of control or arsenic-treated mice as in A were analyzed by anti-Plk1 immunoblotting (IB). The data are presented as means and SD. *, P < 0.05.

Low-dose arsenic induces elevation of Plk1

Considering that Plk1 phosphorylation of HSP70 contributes to attenuation of high-dose arsenic trioxide (5 μM for 24 hours)-induced mitotic abnormalities, such as formation of elongated spindles,^5,10 and that Plk1 elevation leads to aerobic glycolysis,¹³ we then asked whether low-dose arsenic affects the level of Plk1. Toward that end, proteins were prepared from liver, kidney, lung and bladder tissues of mice as described above. As indicated, arsenic-treated mice showed elevated levels of Plk1 in various tissues, suggesting that low-dose arsenic causes elevation of Plk1 in vivo (Fig. 1F). To further confirm this observation, MEFs and human melanoma A375 cells were exposed to low-dose sodium arsenite. The results showed that arsenic treatment significantly increased both mRNA (Fig. 2A) and protein levels of Plk1 (Figs. 2B and C) and that the accumulation of Plk1 protein upon low-dose arsenic treatment were both time- (Fig. 2D) and dose-dependent (Figs. 2E and F). In the time course studies, cells were treated with 0.3 μM sodium arsenite over a 6 hours period. As indicated, we observed a slight increase of Plk1 level as short as 0.5 hour treatment, a significant increase of Plk1 level after 2.5 hours, and the maximum level of Plk1 after 6 hours of treatment (Fig. 2D). The dose response experiments were performed in both MEFs and A375 melanoma cells with 6 hours treatment. While we observed an apparent increase of Plk1 level with 0.3 μM arsenic in MEFs (Fig. 2E), even 0.1 μM arsenic is high enough for us to detect a significant increase of Plk1 level in A375 cells (Fig. 2F). We acknowledge that Plk1 elevation in low-dose arsenic-treated cells (0.3 μM for 6 hours) could be due to cell cycle arrest as high-dose arsenic (5 μM for 24 hours) causes mitotic arrest.⁵ Accordingly, we examined cell cycle status of A375 cells after treatment with 0.3 μM arsenic for 6 hours. As indicated, most of cells remained at G1 phase after such a treatment (Fig. 2G), suggesting that low-dose arsenic-induced Plk1 elevation is not a secondary effect of mitotic arrest.

Figure 2.

Low-dose arsenic induces increased expression of Plk1. (A-C) MEFs were treated with indicated doses of sodium arsenite for 6 h, and harvested for RT-PCR to measure Plk1 mRNA levels (A), anti-Plk1 IB (B) or anti-Plk1 immunofluorescence (IF) staining (C). In A, PDTC is an inhibitor of NF-κB. (D) A375 cells were treated with 0.3 μM sodium arsenite for different times and harvested. (E and F) Arsenic-induced Plk1 elevation is dose dependent. MEFs (E) or A375 cells (F) were treated with different doses of sodium arsenite for 6 h and harvested. (G) Arsenic-induced Plk1 elevation is not due to cell cycle arrest. A375 cells were treated with 0.3 μM sodium arsenite or 50 nM nocodazole (Noc) for 6 h and harvested for IB (left panel) or FACS analysis (right panel).

Low-dose arsenic induces Plk1 elevation in an NF-κB dependent manner

We went on to investigate the mechanism by which Plk1 was induced by low-dose arsenic. Given the previous findings that low-dose arsenic leads to NF-κB activation⁶ and that oxidative stress-induced Plk1 elevation is NF-κB dependent in prostate cells,¹⁴ we next asked whether low-dose arsenic-induced Plk1 elevation is due to activation of NF-κB pathway. As expected, arsenic enhanced the expression of NF-κB (p65 subunit) in a dose-dependent manner with the maximum p65 expression after 1 μM arsenic treatment for 6 hours (Fig. 3A). Of note, no apparent cell cycle arrest was observed under this condition (1 μM for 6 hours) (Fig. 3B). As reported by others,⁵ a significant increase of cell population in G2/M phase was observed at concentrations higher than 2.5 μM (Fig. 3B). To directly test the potential contribution of NF-κB pathway to Plk1 elevation in low-dose arsenic-treated cells, we blocked NF-κB pathway by pyrrolidine dithiocarbamate (PDTC), a specific NF-κB inhibitor.¹⁵ Inhibition of NF-κB pathway indeed reversed low-dose arsenic-induced Plk1 elevation, as shown by both Plk1 mRNA level (Figs. 2A and 3C) and protein level (Fig. 3D). In conclusion, low-dose arsenic-induced Plk1 elevation is due to activation of the NF-κB pathway.

Figure 3.

Low-dose arsenic induces increased expression of Plk1 in an NF-κB-dependent manner. (A) NF-κB is elevated upon arsenic treatment. A375 cells were treated with indicated doses of sodium arsenite for 6 h and harvested for IB. (B) Cell cycle profiles of arsenic-treated cells as in A. (C and D) Inhibition of NF-κB pathway antagonizes arsenic-induced elevation of Plk1. A375 cells were pretreated with PDTC for 1 h, incubated with indicated doses of sodium arsenite for 6 h, and harvested for RT-PCR to measure Plk1 mRNA levels (C) or protein levels (D).

Low-dose arsenic activates the PI3K/AKT/mTOR pathway via Plk1 elevation

Our previous study had revealed that Plk1 is a potent activator for the PI3K/AKT/mTOR pathways by phosphorylating tumor suppressor PTEN.¹³ Given that the PI3K/AKT/mTOR pathway is a central regulator of energy metabolism, we asked whether the PI3K/AKT/mTOR pathway is involved in low-dose arsenic-associated metabolic shift. Indeed, the PI3K/AKT/mTOR pathway was activated by low-dose arsenic treatment, indicated by increased phosphorylation levels of AKT and S6 proteins (Fig. 4A). To determine whether low dose arsenic-induced activation of the PI3K/AKT/mTOR pathway is Plk1 dependent, we first used BI2536, a Plk1 specific inhibitor.¹⁶ Addition of BI2536 clearly reversed low-dose arsenic-induced phosphorylation of AKT (Fig. 4B), suggesting that Plk1 acts upstream of AKT under this condition. To complement the data derived from the use of Plk1 inhibitor, we then employed an RNA interference-based model, in which an inducible knockdown (iKD) of Plk1 in MEFs was achieved by adding doxycycline (Dox).¹⁷ As indicated, an efficient reduction of Plk1 was achieved upon doxycycline treatment (Fig. 4C). More important, knockdown of Plk1 antagonized low-dose arsenic-induced activation of AKT (Fig. 4C), consistent with the inhibitor-based study. Because Plk1 regulates the PI3K/AKT/mTOR pathway via its phosphorylation of PTEN,^13,18 we then monitored phosphorylation status of PTEN upon Plk1 knockdown. As indicated, low-dose arsenic treatment led to elevation of PTEN phosphorylation at both S380 and S385. In contrast, phosphorylation levels of PTEN at S380 and S385 remained pretty low in Plk1-depleted MEFs (Figs. 4D, E), supporting the concept that low-dose arsenic-induced activation of the PI3K/AKT/mTOR pathway is largely due to Plk1-associated PTEN phosphorylation.

Figure 4.

Low-dose arsenic actives the PI3K/AKT/mTOR pathway via induction of Plk1. (A) Arsenic treatment leads to activation of the PI3K/AKT/mTOR pathway. A375 cells were treated with indicated doses of sodium arsenite for 6 h and harvested for IB. (B) Inhibition of Plk1 antagonizes arsenic-induced activation of AKT. A375 cells were pretreated with 50 nM BI2536 for 1 h, incubated with 0.3 μM sodium arsenite for 6 h, and harvested for IB. (C) Plk1 depletion prevents arsenic-induced activation of AKT. MEFs derived from wild-type (WT) or Plk1-iKD mice were pretreated with 50 μg/ml doxycycline for 12 h, incubated with indicated doses of sodium arsenite for 6 h, and harvested for IB. (D and E) Depletion of Plk1 reduces arsenic-induced PTEN phosphorylation. MEFs (WT or Plk1-iKD) were prepared as in C and harvested for IF staining against pS380-PTEN (D) or immunoprecipitation (IP) against PTEN, followed by IB against pS385-PTEN (E).

Arsenic induces cell proliferation and metabolic shift via induction of Plk1

The results described above led us to hypothesize that Plk1 elevation might play a critical role in low-dose arsenic-associated metabolic shift and enhanced cell proliferation. To directly test this hypothesis, we examined the effects of Plk1 deletion on low-dose arsenic-induced cell proliferation and glycolysis. We showed that knockdown of Plk1 attenuated low-dose arsenic-induced elevation of cell proliferation, glucose consumption, and lactate production in MEFs (Figs. 5A–C), arguing that Plk1 indeed plays a critical role in low dose arsenic-induced cell proliferation and glycolysis. To further test whether Plk1 is involved in low-dose arsenic-induced metabolic change in vivo, we performed glucose tolerance test and indirect calorimetry analysis in Plk1-iKD mice. Plk1-iKD mice were pretreated with doxycycline for 2 weeks, followed by 0.01% sodium arsenite treatment. Analysis of Plk1 protein expression in various tissues confirmed efficient Plk1 knockdown upon doxycycline treatment (Fig. 5D). Significantly, no difference was observed in terms of glucose uptake, O₂ consumption, and CO₂ production between arsenic-treated and untreated Plk1 knockdown mice (Figs. 5E–G), supporting that low-dose arsenic-induced metabolism shift is Plk1 dependent in vivo.

Figure 5.

Plk1 is required for low-dose arsenic-induced energy metabolic shift. (A) Low-dose arsenic increases cell proliferation in a Plk1-dependent manner. MEFs (WT or Plk1-iKD) were pretreated with 50 μg/ml doxycycline for 12 h, incubated with 0.3 μM sodium arsenite for 2 or 4 days, and harvested for MTT assay to measure cell viability. (B and C) Depletion of Plk1 antagonizes arsenic-induced aerobic glycolysis. After MEFs as in A were treated with sodium arsenite at 0.3 μM for 24 h, the media were collected for measurement of glucose consumption (B) and lactate production (C). (D) Plk1 knockdown in vivo. Total cellular proteins were prepared from the indicated tissues of doxycycline-treated adult mice (WT or Plk1-iKD) and analyzed by IB. To deplete Plk1, mice were fed with drinking water containing 2 mg/ml doxycycline and 10% sucrose (changed every other day and kept in the dark) for 16 weeks. (E) Plk1 depletion abolishes arsenic-induced increased glucose uptake in mice. Adult Plk1-iKD mice were fed with doxycycline-containing drinking water with or without 0.01% (24 μM) sodium arsenite for 16 weeks, followed by GTT. (F and G) Indirect calorimetry of Plk1-iKD mice treated with low-dose arsenic. Plk1-iKD mice were prepared as in E and subjected to indirect calorimetry for 48 h in metabolic chambers to measure oxygen consumption (F) and carbon dioxide release (G). The data are presented as means and SD. *, P < 0.05; **, P < 0.01.

Inhibition of Plk1 and the PI3K/AKT/mTOR pathway reverses low-dose arsenic-induced glycolysis

Finally, we also used Seahorse XF24 analyzer to detect and quantify glycolytic flux by measuring the extracellular acidification rate (ECAR). For that purpose, cells were growing in glucose-free medium, then treated with glucose, oligomycin, and 2-deoxy-glucose (2-DG) sequentially as indicated (Fig. 6). In this XF Glycolysis Stress Test Assay, glucose addition stimulates ECAR due to increased glycolysis, and addition of oligomycin further enhances ECAR as it inhibits ATP synthase, thus mitochondrial respiration. Finally, addition of 2-DG completely abolishes glycolysis, resulting in the basal level of ECAR. The effect of arsenic on ECAR is dose-dependent. Consistent with the data described above, low-dose arsenic (such as 0.3 μM) apparently boosted glycolytic flux, but high-dose arsenic (such as 5 μM) actually inhibited glycolysis, likely due to severe cytotoxicity (Fig. 6A). Significantly, addition of BI2536 reversed 0.3 μM arsenic-induced increase of glycolytic flux, further supporting a critical role of Plk1 in low-dose arsenic-associated metabolic shift (Fig. 6B). Furthermore, addition of Ly294002, a PI3K inhibitor, and rapamycin, an mTOR inhibitor, also reversed low-dose arsenic-induced increase of glycolytic flux (Fig. 6C and D). These results also support our previous finding that Plk1 elevation-induced glycolysis is likely due to inactivation of PTEN and activation of the PI3K/AKT/mTOR pathway.¹³

Figure 6.

Inhibition of Plk1 reverses low-dose arsenic-induced glycolysis. (A) Low-dose arsenic induces elevation of ECAR. MEFs were treated with arsenic at indicated concentrations for 6 h, followed by glycolytic stress tests using the Seahorse XF24 analyzer. Results were compared using ANOVA (n = 4). (B) Inhibition of Plk1 reverses low-dose arsenic-induced glycolysis. MEFs were treated with arsenic (0.3 µM), BI2536 (50 nM), or a combination of arsenic plus BI2536 for 6 h, followed by glycolytic stress tests. Results were analyzed with ANOVA (n = 6). (C) Inhibition of PI3K reverses low-dose arsenic- induced glycolysis. MEFs were treated with arsenic (0.3 µM), Ly294002 (20 μM), or arsenic plus Ly294002 for 6 h, followed by glycolytic stress tests. Results were analyzed with ANOVA (n = 6). (D) Inhibition of mTOR reverses low-dose arsenic-induced glycolysis. MEFs were treated with arsenic (0.3 µM), rapamycin (50 nM), or arsenic plus rapamycin for 6 h, followed by glycolytic stress tests. Results were analyzed with ANOVA (n = 6).

Discussion

Arsenic is a well-established human carcinogen. Arsenic exposure is presently associated with a spectrum of cancer, including skin, kidney, liver, urinary bladder and respiratory tract cancers.^1-4 Millions of people globally are at risk due to the detrimental effects of arsenic exposure, with its levels in drinking water far exceeding the WHO guideline. Given the consequences, elucidation of the precise molecular mechanism of low-dose arsenic exposure is critical for us to develop appropriate approaches to reduce the potential risk. Arsenic in many ways causes the cell to alter normal signaling pathways. For example, arsenic trioxide inhibits the cell cycle checkpoint activation by taxol, suggesting a potential clinical implication because the efficacy of taxol in the clinic is associated with its ability to induce mitotic arrest and subsequent mitotic catastrophe.¹⁹ There is a tight link between high-dose arsenic-induced apoptosis and mitotic arrest, the latte being one of common mechanisms for arsenic-induced death in cancer cells.²⁰ Moreover, arsenic results in ATP depletion by inhibiting lipoic acid or competing with phosphate in human erythrocytes,²¹ reversed glioblastoma resistance to mTOR-targeted therapies,²² increased oxidative stress due to inhibition of GSH reductase and thioredoxin reductase,²³ alteration of DNA methylation of the promoter of the tumor suppressor gene p53,²⁴ inhibition of DNA repair through regulation of DNA ligase activity,²⁵ and perturbation of multiple signaling pathways, including MAP kinase, p53, AP-1, EIk-1/Egr-1/GADD45α, and NF-κB.^26,27 These analyses suggest that arsenic-associated outcomes are exceedingly complex and that one has to be extremely cautious with data interpretation, in particular, the concentration of arsenic used in individual experiment.

One seminal finding in the arsenic field is that normal tissue might actually benefit from low-dose arsenic during chemotherapy with 5-Fluoroucil (5-FU).⁷ Multiple mechanisms might be responsible for this very intriguing observation. First, low-dose arsenic leads to cytoplasmic accumulation of tumor suppressor p53, thus antagonizing DNA damage-induced p53 response in normal tissues.⁷ The finding described in this manuscript is consistent with this early report. We previously showed that depletion of Plk1 leads to p53 stabilization, suggesting that Plk1 is a negative regulator of p53.²⁸ We later identified 2 p53 regulators, GTSE1 (G2 and S-phase expressed 1) and Topors (topoisomerase 1-binding protein), as Plk1 substrates. We showed that Plk1 phosphorylation of GTSE1 leads to constitutive shuttling of p53 from the nucleus to the cytoplasm, eventually contributing to proteasomal degradation of p53.²⁹ We also showed that Plk1 phosphorylation of Topors inhibits Topors-mediated sumylation of p53 but enhances Topors-mediated p53 degradation.³⁰ Therefore, we propose that low-dose arsenic first leads to elevation of Plk1 and that Plk1-associated phosphorylation of GTSE1 and Topors results in p53 inactivation, eventually protection of normal tissues during chemotherapy. Second, activation of NF-κB and HIF-1α also contributes to low-dose arsenic-associated chemotherapy protection.^6,8 The findings described in this manuscript are consistent with these recent reports as well. We previously showed that oxidative stress-induced elevation of Plk1 is NF-κB dependent in prostate cells¹⁴ and that NF-κB activates Plk1 transcription due to direct binding of RelA subunit to the promoter region of the Plk1 gene.³¹ Therefore, we propose that Plk1 elevation is a major downstream event in response to low-dose arsenic-induced activation of NF-κB. Third, it was shown that low-dose arsenic-associated chemotherapy protection is due to p53/NF-κB-mediated metabolic shift from mitochondrial respiration to glycolysis.⁶ In agreement, we showed that Plk1 elevation also leads to increased glycolysis and decreased oxidative phosphorylation¹³ and that inhibition/depletion of Plk1 reverses low-dose arsenic-induced glycolysis (Figs. 5 and 6). Fourth, we also showed that activation of the PI3K/AKT/mTOR pathway is another downstream event of low-dose arsenic-induced Plk1 elevation (Fig. 4) and that inhibition of the PI3K/AKT/mTOR pathway reverses low-dose arsenic-induced increase of ECAR as well (Fig. 6). This observation is consistent with our previous finding that Plk1 phosphorylation of PTEN results in PTEN inactivation, activation of the PI3K/AKT/mTOR pathway, and a metabolic shift from mitochondrial respiration to glycolysis.¹³ In short, we propose that elevation of Plk1 plays a critical role in low-dose arsenic-induced chemotherapy protection, a concept that is currently under heavy investigation in our laboratory.

Accumulating evidence support the notion that Plk1 has many functions beyond mitosis.¹² For example, we previously showed that Plk1 phosphorylation of Orc2 and Hbo1, 2 members of DNA replication machinery, promotes DNA replication under stressful conditions,^32,33 eventually contributing to acquisition of resistance to drugs that inhibit DNA replication, such as gemcitabine.³⁴ We also showed that Plk1 phosphorylation of CLIP-170 and p150^Glued, 2 microtubule plus-end binding proteins, enhances tubulin dynamics,^35,36 thus contributing to development of resistance to taxol.³⁷ In addition, Plk1 elevation causes activation of androgen de novo biosynthesis pathway, eventually resulting in resistance to androgen receptor inhibitors, such as enzalutamide and abiraterone, 2 major drugs used for patients with castration-resistant prostate cancer.¹⁴ While elevation of Plk1 in cancer cells clearly contributes to therapy resistance, elevation of Plk1 in normal cells due to low-dose arsenic treatment likely contributes to chemotherapy protection. Thus, one has to be very careful when a combination of chemotherapy, low-dose arsenic, and Plk1 inhibitor is used in clinic.

Materials and Methods

Cell culture

MEFs and A375 cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 units/ml streptomycin at 37°C in 8% CO₂.

Antibodies and reagents

The antibodies against PTEN (9188), pS380-PTEN (9551), pAKT (4060), AKT (9272), and pS6 (2211) used in this study were purchased from Cell Signaling. The antibodies against Plk1 (sc-17783) were from Santa Cruz Biotechnology, whereas the antibodies against pS385-PTEN (NG1828769) were from Millipore.

Cell cycle analysis

The standard propidium iodide method was used to determine the cell cycle stage. Briefly, MEFs or A375 cells were cultured in DMEM supplemented with 10% FBS at 37°C in 8% CO2, and treated with sodium arsenite. Upon harvest with 0.1% trypsin, cells were stained with 50 mg/ml propidium iodide in the presence of 100 units/ml RNase A to degrade RNA, followed by FACS (fluorescence activated cell sorting) analysis.

Immunoblotting (IB) and immunoprecipitation (IP) and immunofluorescence (IF) staining

Cell lysates were prepared in TBSN buffer (20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 0.5 mM Na₃VO₄, 5 mM EGTA, 1% NP-40) supplemented with 150 mM NaCl. For IP, lysates were incubated with antibodies in TBSN at 4°C overnight, followed by 3 washes with TBSN plus 500 mM NaCl and 3 additional washes with TBSN plus 150 mM NaCl. For IF staining, cells were plated onto 12-mm glass coverslips in 6-well plates and fixed with 3% paraformaldehyde for 20 min, then followed by permeabilization with 0.1% Triton X-100 for 15 min. Cells were stained with specific antibodies followed by a fluorescein-conjugated secondary antibody (Invitrogen).

Measurement of glucose and lactate

Cells were seeded in culture plates and cultured for 48 hours. The culture media were collected to measure the levels of glucose and lactate with a glucose assay kit (Eton Bioscience) and a lactate assay kit (Eton Bioscience) and values were normalized to the cell number.

Animal experiments

All procedures involving mice were guided by Purdue University Animal Care and Use Committee. Mice were housed in the animal facility with free access to standard rodent chow and water. Mice used in this study were housed under pathogen-free conditions and maintained in a 12-h light/12-h dark cycle. The drinking water that contained 2 mg/ml doxycycline (Sigma, München) and 10% sucrose was prepared every other day (kept in the dark). Groups of 8-week-old mice were provided 0.01% (24 μM) sodium arsenite (Sigma).

Glucose tolerance test (GTT) - GTT was performed by intraperitoneal injection of D-glucose (Sigma) at a dose of 2 mg/g body weight with an Onetouch Ultra glucometer (Lifescan) after an overnight fast. Blood glucose levels were then measured at different times (15 min, 30 min, 45 min, 60 min, 75 min, and 90 min).

Analysis of mouse genotypes

To test the genotype of wild type (WT) or Plk1-inducible knockdown (Plk1-iKD) mice, genomic DNA was prepared from tail clips 0.5–0.8 mm in length with Viagen Direct PCR-Tail reagent (Peqlab Biotechnologie, Erlangen) according to the manufacturer's protocol. For the standard PCR, 10 ng genomic DNA was amplified using the sense primer 5′-ATCGCGGGCCCAGTGTCACTAGGC-3′ and the antisense primer 5′-CTAGTACGCG CCT GCAGGCTAGCC-3′.

Glycolytic flux measurement

Extracellular Acidification Rate (ECAR) was measured with a Seahorse XF24 analyzer by following the manufacturer's protocol. In brief, cells growing in medium without glucose were sequentially treated with 10 mM glucose to stimulate glycolysis, oligomycin to inhibit mitochondrial ATP production and 2-deoxy-glucose (2-DG) to inhibit glycolysis.

Measurement of energy expenditure

Oxygen consumption (VO₂) and carbon dioxide production (VCO₂) were measured under a consistent environmental temperature and light cycle using an indirect calorimetry system (Oxymax, Columbus Instruments). After the mice were acclimated to the metabolic chamber for 2 d, VO₂ and VCO₂ were measured in individual mice at 15 min intervals during a 48 h period.

Statistical analysis

All data are presented as means and standard deviations (SD). Statistical calculations were performed with Microsoft Excel analysis tools. A two-tailed, unpaired Student t test was used to assess the difference between the effects of treatment in cell lines. One-way analysis of variance was used to determine statistically significant differences from the mean in the animal study. P values of <0.05 were considered statistically significant.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by NIH grant R01 CA157429 (X.L.), NIH grant R01 AR059130 (N.A.), NIH grant R01 CA176748 (N.A.), and ACS grant RSG-13–073 (X.L.)