The evaluation of potent antitumor activities of shikonin coumarin-carboxylic acid, PMMB232 through HIF-1α-mediated apoptosis

Biomedicine & Pharmacotherapy 97 (2018) 656–666

Contents lists available at ScienceDirect

Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha

Original article

The evaluation of potent antitumor activities of shikonin coumarincarboxylic acid, PMMB232 through HIF-1α-mediated apoptosis

MARK

Hong-Wei Hana,b, Chao-Sai Zhenga,b, Shu-Juan Chua,b, Wen-Xue Suna,b, Lu-Jing Hana,b, ⁎ ⁎ ⁎ Rong-Wu Yanga,b, Jin-Liang Qia,b, Gui-Hua Lua,b, , Xiao-Ming Wanga,b, , Yong-Hua Yanga,b, a b

State Key Laboratory of Pharmaceutical Biotechnology, Nanjing University, Nanjing 210023, PR China Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry University, Nanjing, 210037, PR China

A R T I C L E I N F O

A B S T R A C T

Keywords: Apoptosis Coumarin HIF-1α Microenvironment Shikonin

In current study, a series of shikonin derivatives were synthesized and its anticancer activity was evaluated. As a result, PMMB232 showed the best antiproliferation activity with an IC50 value of 3.25 ± 0.35 μM. Further, treatment of HeLa cells with a variety of concentrations of target drug resulted in dose-dependent event marked by apoptosis. What’s more, the mitochondrial potential (Δym) analysis was consistent with the apoptosis result. In addition, PARP was involved in the progress of apoptosis revealed by western blotting. To identify the detailed role and mechanism of PMMB232 in the progression of human cervical cancer, we detected the expression of HIF-1α and E-cadherin in HeLa cells. Results showed that expression of HIF-1α was downregulated, while Ecadherin protein was upregulated. Meanwhile, glycolysis related protein PDK1 was decreased in HeLa cells. Conversely, the expression of PDH-E1α was upregulated. Docking simulation results further indicate that PMMB232 could be well bound to HIF-1α. Taken together, our data indicate that compound PMMB232 could be developed as a potential anticancer agent.

1. Introduction Over the past decade, the role of tumor microenvironment, as a pivotal factor influencing tumor resistance, has drawn the attention of researchers. Many components of tumor microenvironment have been identified, which are able to affect tumor sensitivity or resistance to chemotherapy-mediated apoptosis and play a major role in tumor survival and progression [1]. As well, tumors being surrounded by cellular environment is able to form hypoxic niches, which is associated with aberrantly accelerated proliferation of cancer cells. As is known, hypoxia is capable of stabilizing hypoxia-inducible factors (HIFs) which are promptly degraded by an E3-ubiquitin ligase, pVHL, under normoxic conditions [1–5]. As a result, adapting hypoxia promotes various key aspects of cancer progression, causing patients death [6–11]. HIF-1, a heterodimeric protein composed of a HIF-1β subunit in constitutive expression and an O2-regulated HIF-1α subunit, is a major activated transcriptional factor in response to hypoxia [7,12,13]. In addition, its activity is primarily determined by hypoxia with inducing the

stabilization of HIF-1α [11]. It has been reported that HIF-1α is overexpressed in a broad range of human cancer types, and increased levels of HIF-1α activity are often associated with increased tumor aggressiveness and therapeutic resistance [7,10,12]. Given the crucial role of HIF-1α in the association with resistance to drugs in a range of tumors, inhibiting the expression of HIF-1α could be more accurate or more reliable to predict the efficacy of chemotherapy in tumors. Currently, evidence suggest that the activation of HIF-1α signalling pathway induces a vast array of gene products to control energy metabolism, glycolysis, invasion, angiogenesis, apoptosis and cell cycle [14]. As an example, E-cadherin, significant in metastatic colonization, can functionally impact on the changes of metastasis phenotype, which is related to HIF-1α-dependent gene profile involved in angiogenesis and anoikis resistance in breast carcinoma [15]. Rathinasamy Baskaran et al. showed that HIF-1α was required for hypoxia-mediated apoptosis by inhibiting the expression of PARP and up-regulating expression of caspase-3 and caspase-9 [16]. Other studies suggest that HIF-1α is associated with the regulation of p53 in apoptotic neurons [5,17,18].

Abbreviations: HIF-1α, hypoxia-inducible factor alpha 1; DMAP, 4 dimethyamino- pyridine; DCC, N, N’ dicyclohexylcarbodiimide; DMEM, Dulbecco’s modified Eagle’s medium; MTT, 3(4 5-dimethyl-2-thiazolyl)-2 5-diphenyl-2-H-tetrazolium bromide; IC50, Half of maximal inhibitory concentration; SAR, structure–activity relationship; PDB, protein data bank; Δψm, mitochondrial membrane potential; JC-1, 1H-Benzimidazolium 5 6-dichloro-2-[3-(5 6-dichloro-1 3-diethyl-1 3-dihydro-2H-benzimidazol-2-ylidene)-1-propenyl]-1 3-diethyl- iodide; PDK1, 3-phosphoinositide-dependent protein kinase-1; PDH-E1α, pyruvate dehydrogenase (lipoamide) alpha 1; PMSF, phenylmethanesulfonylfluoride; EDTA, ethylene diamine tetraacetic acid; EMT, Epithelial-to-mesenchymal transition; DCFH-DA, 2′,7-dichlorofluoresceindiacetate; ROS, reactive oxygen species ⁎ Corresponding authors at: State Key Laboratory of Pharmaceutical Biotechnology, Nanjing University, Nanjing 210023, PR China E-mail addresses: [email protected] (G.-H. Lu), [email protected] (X.-M. Wang), [email protected] (Y.-H. Yang). http://dx.doi.org/10.1016/j.biopha.2017.10.159 Received 28 June 2017; Received in revised form 30 September 2017; Accepted 28 October 2017 0753-3322/ © 2017 Elsevier Masson SAS. All rights reserved.

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

GAPDH antibody (#E1A7021) and the secondary antibody (anti-mouse or anti-rabbit IgG) (#E1WP319 or #E1WP318) were purchased from Enogene (Nanjing, China). ECL Kit (#34077) was purchased from Thermo Scientific (USA). 3-(4, 5- Dimethylthiazol-2-yl) −2, 5-diphenyltetrazolium (MTT) were from Beyotime Institute of Biotechnology (Haimen, China). Fibronectin (#F1056) and laminin (#L2020) were purchased from Sigma–Aldrich (St. Louis, MO).

Energy production is linked intimately with almost all cellular events. Thus, targeting at HIF-1α to modulate hepatic glucose metabolism may provide a promising therapy against energy metabolism with the relation to diseases, such as diabetes, postsurgical liver dysfunction and even cancers [10,19,20]. Shikonin and its derivatives are active naphthoquinone compounds extracted from the root of a Chinese herbal medicine Lithospermum erythrorhizon [21–24]. Currently, numerous evidences suggest that shikonin induces apoptosis process by p53 directed mitochondrial pathway, and suppress the expression of the family members of anti-apoptotic Bcl-2 (B-cell lymphoma 2). Meanwhile, it also increases the activities of caspases, inactivates AKT pathway, induces cancer cell apoptosis [25–28]. These results, together with our earlier findings, indicate that shikonin may have high efficacy for preventing and treating cervical cancer in the future. However, there are not many investigations on its precise anticancer effect and mechanism of inducing apoptosis in cervical cancer cells by reducing HIF1α. It will be important in the future to ask whether the effort of shikonin is efficient to inhibit hypoxia-inducible factors and break the functions of mitochondria. Ultimately, disrupting cellular energy metabolism leads to cells apoptosis. Whereas, owing to poor solubility and gigantic cytotoxic effects on non-cancer cells, it is hindered to develop as a new clinical anticancer agent. More importantly, a lot of coumarin compounds as medicinal candidates with strong pharmacological activity, low toxicity and side-effect, little drug resistance, high bioavailability, broad spectrum and good curative effects were being actively developed [29–32]. Recent studies also highlighted that newly synthesized coumarin substituted derivatives showed good water- solubility [33–35]. This could be attributed to the formation of hydrogen bonds between coumarin groups and water molecules [33]. In addition, curcumin was considered to interrupt HIF-1α expression at protein levels, promote HIF-1α degradation and achieve anticancer agent potential by targeting at HIF-1α [36–38]. Based on the aforementioned results, we proposed that coumarin into shikonin by introducing aminobenzoic acids as bridges may improve its water solubility and cervical tumor targeting.

2.2.1. Cell lines and culture conditions Human cervical cell lines (HeLa), human lung adenocarcinoma epithelial cell line (A549), human breast cancer cell line (MCF-7, MDAMB-231), human embryonic kidney cells (293T) and L0-2, human normal liver cell lines, were obtained from American Type Culture Collection (ATCC; Man-assas, USA). All cells were maintained in DMEM medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Hyclone, China), 100 U/mL penicillin and 100 μg mL−1 streptomycin (Sigma, Brazil) at 37 °C and 5% CO2 at the beginning of each experiment. 2.2.2. MTT assay for cell viability The cell viability of cancer cells was determined by MTT assay. Briefly, cells were seeded in 96-well plates at a density of 2 × 104 cells/ well and then allowed to adhere for 24 h at 37 °C in a humidified atmosphere with 5% CO2. After that, cells were treated with various concentrations of test compounds (0.1, 1, 10 and 100 μM) with shikonin as positive reference. Following a 24 h incubation, a volume of 20 μL of PBS containing 4 mg mL−1 of MTT was added to each well. Besides, plates were incubated for a further 4 h, before they were centrifuged at 1500 rpm at 4 °C for 10 min, followed by the removal of the supernatant. Then, DMSO (150 μL) was added to each well for coloration, and the plates were subsequently shaken vigorously to ensure complete solubilization for 10 min at room temperature. The light absorption (OD, optical densities) were recorded on a BioTek ELx800 ELISA reader at a test wavelength of 570 nm and a reference wavelength of 650 nm. In all experiments, three replicate wells were used only for each drug concentration, while each assay was performed for at least three times. The effect of PMMB229-240 on tumor cell viability was expressed by IC50 of each cell line and the results are presented in Table 2.

2. Materials and methods 2.1. Chemistry All chemicals and reagents used in current study were analytical grade. 4-hydroxycoumarin was purchased from J & K SCIENTIFIC LTD (Shanghai, China). Shikonin, doxorubicin and all acids were purchased from Sigma Aldrich (St. Louis, MO). All the 1H NMR spectra were recorded on a Bruker DPX 300 Spectrometer in CDCl3 and chemical shifts (δ) were reported as parts per million (ppm). ESI–MS spectra were recorded a Mariner System 5304 Mass spectrometer. Melting points were determined on a XT4 MP apparatus (Taike Corp, Beijing, China). Thin layer chromatography (TLC) was performed on silica gel plates (Silica Gel 60 GF254) and visualized in UV light (254 nm). Optical rotations were measured with a Rudolph Research Analytical II automatic polarimeter, using a sodium lamp (589 nm). The general procedure for preparation of compounds A229-A240, B229-B240 and PMMB229PMMB240 were appended in the Supplementary data (S1.1).

2.2.3. Cell apoptosis analysis Cells were plated in 6-well plates (2 × 105 cells per well) and incubated at 37 °C for 24 h. Fresh medium containing a variety of concentrations of PMMB232 were then added to culture dishes. Following 24 h incubation, the untreated cells (control) or cells treated were centrifuged with (2500 rpm at 4 °C for 10 min), rinsed twice with precooled phosphate buffered saline (PBS), and prepared for apoptosis analysis. Cells were stained with FITC-conjugated Annexin V-FITC incubation for 10 min and then performed propidium iodide stain for another 10 min at room temperature in dark (Life Technologies). Flow cytometric analysis was conducted after supravital stain immediately. 2.2.4. Cell adhesion assay Cells adhesion assay was performed using the method published by Lin et al. (Lin et al., 2015) with some modifications. Briefly, 96-well flat-bottom plates were coated with 50 uL fibronectin and laminin (10 ug mL−1) in PBS overnight at 4 °C and then blocked with 0.2% BSA for 2 h at room temperature followed by three times washing. Then, HeLa cells, which had been presented by PMMB232, shikonin or doxorubicin for 24 h, were added to each well (2 × 105 per well) in triplicate, and incubated at 37 °C, 5% CO2 for 40 min. Next, plates were washed twice with PBS to eliminate unbound cells. Cells remained adhering to the plated were determined by MTT assay. The binding cells were calculated by dividing the optical density of initial input cells. Each assay was carried out at least three times.

2.2. Biology In our study, we used the following materials: Annexin V-FITC cell apoptosis assay kit (#BA11100) was purchased from BIO-BOX (Nanjing, China). BCA protein assay kit (#23227) was available from Pierce (Rockford, IL, USA). PVDF membranes were recruited from Biosharp (Hefei, China). Anti-HIF-1α (#WL01607) and anti-E-Cadherin (#WL01482) were purchased from Wanleibio. (Shenyang, China) AntiPDH-E1α Antibody (D-6) (#sc-377092), anti-p-PDH-E1α (#ab177461), anti-PARP (#G3014) and cleaved PARP (#I1013) antibody were purchased from Santa Cruz Biotechnology, inc. Anti-PDK1 (#E1A6018-1), 657

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Scheme 1. The routes to synthesizing the novel coumarin-carboxylic shikonin ester derivatives PMMB229-240.

2.2.7. Measurement of mitochondrial membrane potential using JC-1 staining Cells were seeded on 6-well plate at a density of 1 × 105 cells/well in 2 mL medium in 10% FBS. After incubation for 24 h, the cells were treated with various concentrations of PMMB232 (4 and 8 μM) for 12 h. Cells in each group were collected after trypsinization with 0.25% trypsin and were incubated with 1 mL JC-1 dye (4 mL distilled water, 1 mL dyeing buffer and 25 μL dyeing) (Invitrogen) at 37 °C and 5% CO2 for 20 min. Then, cells were centrifuged, washed with dyeing buffer. The mitochondrial membrane potential was assessed by analyzing the red and green fluorescence after using flow cytometer analysis via fluorescence-activated cell sorting BD (FACS) caliber. (BD, USA).

2.2.5. Wound healing assays Cells were seeded in a tissue culture 6-well plate at an initial density of 2.4 × 105 cells/cm2 overnight. A micropipette tip was used to create a wound in the monolayer by scraping after cells seeding and incubation for 24 h. Identical cell-free space was observed by phase-contrast microscopy (NIKON, Japan) and digital images were taken by cells at 0 h, 12 h and 24 h. Subsequently, the NIH Image J image analysis software was used to outline the wound areas and analysis quantifies the reduced cell-free space, n = 3. 2.2.6. Measurement of intracellular ROS level Intracellular ROS were detected by using the total ROS detection kit (Beyotime Institute of Biotechnology, China) according to the manufacturer’s instructions. Briefly, cells were seeded in a 12-well plate at approximately 60%–70% confluence. After indicated treatment, cells were washed with PBS and were either stained directly with ROS detection solution at 37 °C for 20 min, washed in serum-free medium thrice then analyzed by using a flow cytometer (BD Biosciences) or by quantitative measurement of the fluorescence intensity of the cells exposed to PMMB232 or positive control observed under an Olympus confocal microscope and data was analyzed using FV-10-ASW 1.7 viewer.

2.2.8. Western blot analysis HeLa cells were treated with PMMB232 (2, 4, or 8 μM) for 24 h. For protein isolation, medium was removed, cells were washed twice with ice-cold PBS, then lysed using cell lysis buffer [50 mmol L−1 Tris (pH 7.4), 150 mmol L−1 NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, and sodium orthovanadate, sodiumfluoride, EDTA, leupeptin, 1 mmol L−1 phenylmethanesulfonylfluoride (PMSF)]. The lysates were col-lected by scraping from the plates and then centrifuged

Table 1 The structure of the compound PMMB229-240.

Compound

R

Compound

R

Compound

PMMB229

PMMB233

PMMB237

PMMB230

PMMB234

PMMB238

PMMB231

PMMB235

PMMB239

PMMB232

PMMB236

PMMB240

658

R

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Table 2 The cytotoxicity of compound PMMB229-240 against a panel of human cancer cell lines and two non-cancer cell line. Compounds

PMMB229 PMMB230 PMMB231 PMMB232 PMMB233 PMMB234 PMMB235 PMMB236 PMMB237 PMMB238 PMMB239 PMMB240 shikonin doxorubicin

IC50(μM) Hela

MDA-MB-231

MCF-7

A549

293T

L02

15.41 ± 1.12 11.20 ± 1.01 10.02 ± 3.21 3.25 ± 0.35 19.36 ± 2.24 18.67 ± 2.01 12.22 ± 3.76 16.00 ± 3.89 19.18 ± 2.30 21.08 ± 2.40 19.18 ± 2.30 21.08 ± 2.40 6.50 ± 1.89 4.61 ± 0.36

18.85 ± 2.16 11.35 ± 2.39 13.06 ± 2.05 9.73 ± 3.02 27.89 ± 3.29 16.45 ± 4.01 26.35 ± 3.19 20.98 ± 2.88 13.02 ± 1.09 18.56 ± 2.37 13.02 ± 1.09 18.56 ± 2.37 8.98 ± 1.12 4.21 ± 1.09

27.77 ± 4.99 22.28 ± 2.10 21.15 ± 2.07 10.56 ± 1.00 26.23 ± 1.58 32.75 ± 3.03 25.92 ± 2.51 29.51 ± 4.19 26.68 ± 1.79 25.72 ± 1.39 26.68 ± 1.79 25.72 ± 1.39 7.75 ± 1.32 3.97 ± 1.02

21.90 ± 1.91 16.78 ± 1.78 25.70 ± 5.40 8.73 ± 2.93 17.44 ± 0.87 23.76 ± 3.26 12.05 ± 1.15 24.18 ± 4.08 26.48 ± 2.48 45.97 ± 5.07 26.48 ± 2.48 45.97 ± 5.07 10.37 ± 1.73 5.13 ± 0.78

59.32 ± 3.92 44.87 ± 4.87 59.54 ± 5.04 63.80 ± 3.80 51.57 ± 3.19 56.96 ± 3.69 61.16 ± 3.16 64.43 ± 4.43 55.43 ± 5.33 53.34 ± 3.35 55.43 ± 5.33 53.34 ± 3.35 1.52 ± 0.45 3.77 ± 0.62

43.78 ± 3.89 55.63 ± 4.32 67.10 ± 5.01 35.04 ± 2.80 56.57 ± 4.19 39.46 ± 2.97 71.61 ± 6.31 44.39 ± 2.67 75.34 ± 5.17 83.64 ± 7.05 75.34 ± 5.17 83.64 ± 7.05 6.22 ± 1.31 1.09 ± 0.27

at 10, 000 rpm at 4 °C for 10 min. Total protein samples (80 μg) were loaded on 10% of SDS-polyacrylamide gels for electrophoresis, and then transferred onto PVDF transfer membranes (Millipore, Billerica, USA) at 200 mA for 2 h. Membranes were blocked at room temperature for 1 h with 5% nonfat milk in TBS-Tween buffer (0.12 M Tris–base, 1.5 M NaCl, 0.1% Tween 20) for 1 h at room temperature. Membranes were then incubated overnight at 4 °C with primary antibodies (anti-PDK1, anti-PDH-E1α, anti-p-PDH-E1α, anti-HIF-1α, anti-E-Cadherin, antiPARP, anti-cleaved PARP and anti-GAPDH) at a dilution of 1:500

(Biosynthesis Biotechnology Company, Beijing, China) in blocking solution. After thrice washings in TBST for each 5 min, membranes were incubated for 1 h at room temperature with alkaline phosphatase peroxidase-conjugated secondary antibody (1:5000 dilution) in blocking solution. Detection was performed by the BCIP/NBT Alkaline Phosphatase Color Development kit (Beyotime Institute of Biotechnology) according to the manufacturer’s instructions. Bands were then recorded by a digital camera (Tanno, China).

Fig. 1. Effects of PMMB232, shikonin and B232 on cell apoptosis in HeLa and L02 cells. HeLa cells were all treated with 0, 2, 4 μM PMMB232, shikonin (4 μM) and B232 (8 μM) for 24 h. L02 cells were treated with 0, 4, 8 μM PMMB232, shikonin (4 μM) and B232 (8 μM) for 24 h (A) PMMB232 showed pro-apoptotic effect in a dose-dependent manner on HeLa cells, but (B) had only slight effects on cell apoptosis of L02 cells.

659

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Fig. 2. Effect of PMMB232 and shikonin on the mitochondrial membrane potential of HeLa and L02 cells after 12 h. (A) Effect of PMMB232 on the mitochondrial membrane potential of HeLa cells (B) Effect of PMMB232 and shikonin on the mitochondrial membrane potential of L02 cells Data are representative of three independent experiments. Significant differences (*P < 0.05 or **P < 0.01) between PMMB232-treated and untreated control cells.

3.2. Biological evaluation

2.2.9. Docking simulation The refined protein molecules (PMMB232 for HIF-1α (PDB code: 4W9I)) were used, and the docking methodology was performed according to a previously reported protocol [39].

3.2.1. PMMB232 inhibits the proliferation of HeLa cells All the newly synthesized compounds of PMMB229-240 were evaluated in vitro growth inhibitory activities against four cancer cell lines. These lines included human cervical cell line (HeLa), human lung adenocarcinoma epithelial cell line (A549), human breast cancer cell line (MCF-7, MDA-MB-231) and two non-cancer cell lines of human embryonic kidney cells (293T) and human normal liver cells (L0-2) with doxorubicin and shikonin as references. Half of maximal inhibitory concentration (IC50) values were obtained from an inhibitory model with the sum of squares of residuals minimized by Origin 7.5 software. As shown in Table 2, some of target compounds could effectively inhibit the proliferation of four tumor cells, but their effects were not better than that of shikonin or doxorubicin. However, the introduction to coumarin-carboxylic esters greatly reduced the cytotoxicity of shikonin towards normal cells instead of avoiding cancer cells. For example, PMMB230 and PMMB231 selectively inhibited the proliferation of MDA-MB-231 with an IC50 value of 11.35 ± 2.39 μM and 13.06 ± 2.05 μM, which didn’t have significant cytotoxicity towards 293T (44.87 ± 4.87 μM, 59.54 ± 5.04 μM) and L0-2 (55.63 ± 4.32 μM, 67.10 ± 5.01 μM). PMMB235 (12.05 ± 1.15 μM) also showed comparable inhibition effect against the shikonin (10.37 ± 1.73 μM) in A549 cells. What’s more,

3. Results 3.1. Chemistry 3.1.1. General procedure for preparation of compounds PMMB229PMMB240 In our present work, a series of novel coumarin-carboxylic shikonin ester derivatives, PMMB229-PMMB240, were synthesized (Scheme 1). These synthetic compounds were presented in Table 1. All of these synthetic compounds offered satisfactory elementary analyses and spectroscopic data. Results obtained by proton nuclear magnetic resonance (1H NMR) and electrospray ionization-mass spectroscopy (ESI–MS) were consistent with assigned structures (in the Supplementary data S2).

660

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Fig. 3. The influence of PMMB232 on the ROS level of HeLa cells. (A) DCF fluorescence was detected by confocal microscope after cells treated with PMMB232 (4 μM) or Rosup (50 μg/mL) alone for 12 h. (B) Bars represent generation of intracellular ROS in HeLa cells. Data shown are the mean ± S.E.M. of three independent experiments. *P < 0.05, **P < 0.01. (C) DCF fluorescence was detected by flow cytometer after cells treated with PMMB232 (4 μM) or Rosup alone for 12 h. (D) DCF fluorescence was detected by flow cytometer after L02 cells treated with PMMB232 (8 μM) and HeLa cells treated with PMMB232 (4 μM) or Rosup alone for 12 h (the curve shifts to the right when compared with the L02-treatment groups).

PMMB232 (12.5%) under the same conditions, or even with the exposure concentration extended to 8 μM. But, the apoptosis inducing the effect of shikonin (31.4%) on L0-2 cells was more significant than that of PMMB232 (12.5%) and B232 (4.3%) (Fig. 1B). Above all, PMMB232 induced an apoptosis ratio greater than that of shikonin efficacy, while the toxicity of drugs in non-cancerous cells (L0-2) was less than that of shikonin.

PMMB232 exhibited the most comprehensive anti-proliferation effect against the four cancer cell lines, and displayed the most potent cytotoxicity towards HeLa cells with an IC50 value of 3.25 ± 0.35 μM, which was comparative to the positively controlled shikonin (IC50 = 6.50 ± 1.89 μM) and doxorubicin (IC50 = 4.61 ± 0.36 μM). 3.2.2. PMMB232 induce apoptosis in HeLa and L0-2 cells To further ascertain whether the anti-proliferative activity of PMMB232 induced cells apoptosis, we analyzed the proportion of apoptotic cells after treating various concentrations of PMMB232, B232 and shikonin. The results revealed the effect of PMMB232 on cell apoptosis of HeLa and showed that L0-2 cell lines were different. For HeLa cells treated with PMMB232, the proportion of apoptotic cells was 38.0%, significantly higher (P < 0.01) than that of the group dealt with shikonin (20.8%) or B232 (34.3%) (Fig. 1A). However, there were no significant changes in the apoptotic cells of L0-2 cells treated with

3.2.3. Effect of PMMB232 on the mitochondrial membrane potential (Δψm) of HeLa cells Maintenance of mitochondrial membrane potential was fundamental for the performance and survival of cells [40]. To further confirm cell apoptosis because of PMMB232, the degree of mitochondrial damage in terms of the depolarization of ΔΨm was measured by flow cytometer with JC-1. As shown in Fig. 2A, a dose-dependent loss of ΔΨm was observed in HeLa cells after the challenge of cells with 661

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Fig. 4. Influence of PMMB232, shikonin and B232 on HeLa cell adhesive to fibronectin and laminin. (A) Influence of PMMB232, shikonin and B232 on HeLa cells adhesive to fibronectin (B) Influence of PMMB232, shikonin and B232 on HeLa cells adhesive to laminin. Data shown are the mean ± S.E.M. of three independent experiments. *P < 0.05, **P < 0.01.

Fig. 5. Cell migration was detected by the wound scrape assay. Representative images of cell migration in the wound scrape model at 0 h, 12 h and 24 h are shown. The data represent the means ± SD of three experiments. *P < 0.05, **P < 0.01.

662

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Interestingly, the production of ROS was significantly higher in HeLa cells than in L0-2 cells. It was notable that PMMB232 could enhance HeLa cells growth inhibition by inducing ROS accumulation with negligible toxic effects to non-cancer cells. 3.2.5. PMMB232 induce the adhesive ability of HeLa cells Epithelial-to-mesenchymal transition (EMT) was a key feature of invasive cells and could be characterized by the loss of epithelial cellcell contact and the acquisition of mesenchymal features and motility. Herein, the adhesive ability of HeLa cells influenced by PMMB232, B232 and shikonin in this assay was examined (Fig. 4A and B). These data showed that the cell adhesion of drug-treated group decreased 1.2 times, compared with the control group, but the group treated with shikonin slightly decreased. As a whole, PMMB232 could significantly reduce the adhesive ability of HeLa cells to fibronectin and laminin, but its effects on shikonin and B232 were not obvious. 3.2.6. Wound-healing migration assay Tumor cell migration was identified as one hallmark of malignant tumor progression, particularly of the establishment of lethal secondary metastases at distant organs. To measure the migratory capacity of cells, the conventions in vitro wound healing assay were conducted. Fig. 5 showed once the pressure (0 h) was released, a wound (width: 250 μm, length: 4 mm) with a clear edge was formed between cellmonolayers. Then, migration was characterized by photographing cellfree area during the following 0 h, 12 h and 24 h. The archetypal images (Fig. 5A) showed that the cells gradually occupied the cell-free space within the channel. The particle analysis of Image J was applied to quantify cell-covered area (Fig. 5B). These results showed that the reduction of the cell-free space was separate 23.4%, 39.1% and 77.2% after 0 h, 12 h and 24 h serum-drove cell migration [Image J particle analysis quantifies the reduced cell-free space, n = 3].

Fig. 6. Relative expression of HIF-1α and E-cadherin protein in HeLa cell lines was measured by Western blot. Images are representative of three independent experiments. *P < 0.05, **P < 0.01. Blots were reprobed for GAH as a loading control.

PMMB232, as evidenced by an increase in the cell number with high JC-1 green fluorescence. From contour plots, it was clear that the effect of PMMB232 on the depolarization of ΔΨm was stronger than that in the control group. The percentage of the cells with depolarized ΔΨm (within the selected region defined by green area) increased from 12.6% of the control group to 20.3% (2 μM) or 63.1% (4 μM) of the HeLa cells treated with PMMB232, and to 55.9% in the positive control (4 μM shikonin) conditions. At the same time, the loss of mitochondrial membrane potential of L0-2 cells was measured. Results in Fig. 2B implied that JC-1 red fluorescence loss rate of L0-2 cells was 7.99% ± 1.01% by PMMB232 (2 μM), 19.00% ± 3.20% by PMMB232 (4 μM), and 56.6% ± 5.17% by shikonin (4 μM). These observations indicated that the effect of PMMB232 on the mitochondrial membrane potential of HeLa cells was better than that of shikonin with minimal toxic effect to non-cancer cells under the same conditions.

3.2.7. Expression of HIF-1α and E-cadherin in HeLa cells Solid tumors have very unique physiological characteristics such as abnormal vasculature and intra-tumor hypoxia [5]. To protect against cell death due to oxygen depletion, cancer cells stabilized HIF-1 to induce the expression of numerous target genes related to cell growth, metabolism and invasion [4,41–43]. Thus, to investigate the roles of HIF-1α and E-cadherin in the progression of cervical cancer, we firstly detected the expression of HIF-1α and E-cadherin proteins in HeLa cells. In the Western Bolt results shown in Fig. 6, the expression of HIF1α decreased 3-fold in drug-treatment cells in comparison to the negative control cells. We also found an increase of 1.5-fold in drugtreatment group than in the control group. While, both of the two proteins do not show any significant changes in L0-2 cells (Fig. S1). It suggested that inhibiting the expression of HIF-1α weakened glycolysis and apoptosis-related activities in cancer cells rather than in non-cancer cells. In contrast, the up-regulation of E-cadherin enhanced the adhesion between cells, inhibiting tumor cell migration and metastasis.

3.2.4. PMMB232 induces oxidative stress in HeLa cells Excessive production of ROS in cells was known to induce apoptosis. The ability of ROS to inflict severe cellular damage and cell death had been exploited as an approach to kill cancer cells. Thus, intracellular ROS generation was evaluated by intracellular peroxide-dependent oxidation of 2′, 7-dichlorofluoresceindiacetate (DCFH-DA) to form fluorescent DCF. DCF fluorescence was detected after cells treated with PMMB232 or Rosup (Reactive Oxygen Species positive control reagent) alone for 12 h. The production of ROS was strikingly increased in the treatment of PMMB232, compared with the control group. (P < 0.01) For instance, the ROS level was as 1.86 times as the control group with 4 μM cells treated with PMMB232 (Fig. 3A). However, for positive control, the ROS level was as 1.63 times as the control group, much lower than that of the cells treated with PMMB232 under the same conditions. To conform this important finding, the L0-2 cells treated with PMMB232 were examined. As shown in Fig. 3B, the production of ROS was slightly changed, compared with the control group.

3.2.8. Expression of PDK1 and PDH-E1α in HeLa cells HIF-1α family members were known as key regulators of glycolysis. PDK1 was a hypoxia- responsive protein that actively regulated the functions of mitochondria under hypoxic conditions by reducing pyruvate into TCA cycle [13,44–48]. Hence, the effects of shikonin derivatives on the expression levels of PDK1 and PDH-E1α in HeLa cells were examined by Western Blot. As shown in Fig. 7, mild increase of the expression of PDH-E1α was observed in cells treated with PMMB232. However, the drug-treatment significantly suppressed the expression levels of PDK1 and decreased the expression levels of p-PDH-E1α. It indicated that PMMB232 could induce more pyruvate into TCA cycle instead of Warburg effect. However, this phenomena do not occur in L02 cells (Fig. S2). Accumulating evidences highlighted E-cadherin as one of the fundamental factors in the progression of cancer cell proliferation, invasion and metastasis [49,50]. 663

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Fig. 8. HeLa cells were treated with PMMB232 (0, 4, 8 μM and shikonin (8 μM)) and the full-PARP and cleaved PARP protein level were analyzed by western blotting and GAPDH as a loading control. Images are representative of three independent experiments. *P < 0.05, **P < 0.01.

Fig. 7. Western blots of HeLa cells lysates after exposure to the PMMB232 concentrations for 24 h. Blots were probed for PDK1, PDH-E1α (D-6), PDH-E1α (p-Ser293) and GAPDH as a loading control. Images are representative of three independent experiments. *P < 0.05, **P < 0.01.

4. Discussion Series of the shikonin derivatives of PMMB229-PMMB240 were designed. Among these novel compounds, PMMB232 had the superior efficacy for cancer chemo-preventive activities. At the same time, the treatment of HeLa cells with a variety of concentrations of PMMB232 resulted in a dose-dependent event marked by apoptosis. In harmony with apoptosis results, mitochondrial potential (Δψm) analyses showed that the effect of PMMB232 on the mitochondrial membrane potential of HeLa cells was better than that of shikonin with minimal toxic effect to non-cancer cells under the same conditions. ROS generation was shown to play an important role in apoptosis. In 12 target compounds, PMMB232 had unique efficacy to result in increased levels of intracellular ROS in HeLa cells. Furthermore, the results of woundhealing migration assay and the adhesive ability of HeLa cells assay all confirmed the potential anti-cancer capacity of PMMB232. To address the detailed roles and mechanism of PMMB232 in the progression of cervical cancer, we detected the expression of HIF-1α, E-cadherin and glycolytic proteins in HeLa cells [47,49]. Results showed that the target drugs could induce the degradation of HIF-1α, which induced the expression of numerous target genes related to cell growth and metabolism [44]. However, the expression of E-cadherin, as a decisive event underlying the epithelial-to-mesenchymal transition in cancer cells, was increased under the treatment of PMMB232. Meanwhile, PDK1 was a very potent inhibitor of PDH complex and the expression level of PDH was marginal in up-regulation. It indicated that PMMB232 could induce more pyruvate into TCA cycle instead of Warburg effect [41]. In addition, our results showed that the administration of PMMB232 for 24 h induced a significant dose-dependent down-regulation of the apoptotic protein levels of PARP, while cleaved-PARP was upregulated in HeLa cells. Our findings suggested that PMMB232 could be bound to HIF-1α, prompting the degradation of HIF-1α and weakening glycolysis and apoptosis-related activities in cancer cells. What is more, the CDOCK interaction energy values of compounds bound to HIF-1α (PDB

3.2.9. Expression of PARP and cleaved PARP in HeLa cells To further confirm the major role of HIF-1α involving apoptosis pathway, the expression levels of PARP and cleaved PARP were detected (Fig. 8). As expected, the expression of cleaved PARP was remarkably increased and no obvious changes of PARP levels were observed. The expression of PARP and cleaved PARP were all no obvious changes in L0-2 cells yet (Fig. S3). 3.2.10. Molecular docking of PMMB232 To better understand the potency of PMMB232 and further guide the structure–activity relationship (SAR) studies, we examined the interaction of PMMB232 with HIF-1α (PDB code: 4W9I). All docking runs were applied with the Lamarckian genetic algorithm of Auto-Dock 3.5. All amino acid residues which interacted with HIF-1α were exhibited. The interaction of PMMB232 with HIF-1α amino acid residues was depicted in Fig. 9A. In binding model, PMMB232 was nicely bound to the binding site of HIF-1α via four hydrogen bonds with His 115 (distance = 2.9437 Å), TYR 112 (distance = 2.9909 Å), ARG (distance = 3.0204 Å) and two π bonds of TYR 112 (distance = 5.9441 Å) and TYR 98 (distance = 5.1457 Å). Fig. 9B depicted the 3D models of the interaction between PMMB232 and HIF-1α. The results of molecular docking argued that PMMB232 may be a potential anti-HIF-1α agent. Also, the CDocker Energy (energy of ligand-receptor complexes- kcal/mol) of 12 compounds was also depicted in Fig. S4, which agreed with the trends of HIF-1α inhibitory and anti-proliferative inhibitory. Furthermore, CDOCK interaction energy values (table S1) of other derivatives (with different R substituents), lower than that of PMMB232, excluded the 12 compounds, mentioned above, bound to HIF-1α. This was probably due to steric hindrance or electron donating effect that prevented the linkage 664

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

Competing interests The authors have declared that no competing interests exist. Acknowledgements This research was supported by the Program for Changjiang Scholars and Innovative Research Team in University (IRT_14R27), the National Natural Science Foundation of China (31470384, 31171161, and 31670298), and the Fundamental Research Funds for the Central Universities (020814380002) Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.biopha.2017.10.159. References [1] K. Yaacoub, R. Pedeux, K. Tarte, T. Guillaudeux, Role of the tumor microenvironment in regulating apoptosis and cancer progression, Cancer Lett. 378 (2016) 150–159. [2] D.H. Bach, S.H. Kim, J.Y. Hong, H.J. Park, D.C. Oh, S.K. Lee, Salternamide a suppresses hypoxia-Induced accumulation of HIF-1α and induces apoptosis in human colorectal cancer cells, Mar. Drugs 13 (2015) 6962–6976. [3] R. Yin, L. Yuan, L.L. Ping, L.Y. Hu, Neonatal bronchopulmonary dysplasia increases neuronal apoptosisin the hippocampus through the HIF-1α and p53 pathways, Respir. Physiol. Neuro. 220 (2016) 81–87. [4] C. Furuta, T. Miyamoto, T. Takagi, Y. Noguchi, J. Kaneko, S. Itoh, T. Watanabe, F. Itoh, Transforming growth factor-beta signaling enhancement by long-term exposure to hypoxia in a tumor microenvironment composed of Lewis lung carcinoma cells, Cancer Sci. 106 (2015) 1524–1533. [5] H. Yasuda, Solid tumor physiology and hypoxia-induced chemo/radio-resistance: novel strategy for cancer therapy: nitric oxide donor as a therapeutic enhancer, Nitric Oxide 19 (2008) 205–216. [6] R. Zhang, X.J. Song, C. Liang, G.S. Song, Y. Chao, Y. Yang, K. Yang, L.Z. Feng, Z. Liu, Catalase-loaded cisplatin-prodrug-constructed liposomes to overcome tumor hypoxia for enhanced chemo-radiotherapy of cancer, Biomaterials 138 (2017) 13–21. [7] G.L. Semenza, Defining the role of hypoxia-inducible factor 1 in cancer biology and therapeutics, Oncogene 29 (2010) 625–634. [8] Y.C. Chang, Y.C. Chan, W.M. Chang, Y.F. Lin, C.J. Yang, C.Y. Su, M.S. Huang, A.T.H. Wu, M. Hsiao, Feedback regulation of ALDOA activates the HIF-1α/MMP9 axis to promote lung cancer progression, Cancer Lett. 403 (2017) 28–36. [9] Q. Zhang, Y. Lou, J.Y. Zhang, Q.H. Fu, T. Wei, X. Sun, Q. Chen, J.Q. Yang, X.L. Bai, T.B. Liang, Hypoxia-inducible factor-2α promotes tumor progression and has crosstalk with Wnt/β-catenin signaling in pancreatic cancer, Mol. Cancer 16 (2017) 1–14. [10] J. Chen, N. Imanaka, J. Chen, J.D. Griffin, Hypoxia potentiates Notch signaling in breast cancer leading to decreased E-cadherin expression and increased cell migration and invasion, Br. J. Cancer 102 (2010) 351–360. [11] G.L. Semenza, Hypoxia-inducible factors in physiology and medicine, Cell 148 (2012) 399–408. [12] N. Liu, W.Y. Xia, S.S. Liu, H.Y. Chen, L. Sun, M.Y. Liu, L.F. Li, H.M. Lu, Y.J. Fu, P. Wang, H.L. Wu, J.X. Gao, MicroRNA-101 targets von Hippel-Lindau tumor suppressor (VHL) to induce HIF1 alpha mediated apoptosis and cell cycle arrest in normoxia condition, Sci. Rep. 6 (2016) 1–13. [13] Z.H. Ai, Y. Lu, S.B. Qiu, Z. Fan, Overcoming cisplatin resistance of ovarian cancer cells by targeting HIF-1-regulated cancer metabolism, Cancer Lett. 373 (2016) 36–44. [14] B. Keith, R.S. Johnson, M.C. Simon, HIF1α and HIF2α: sibling rivalry in hypoxic tumour growth and progression, Nat. Rev. Cancer 12 (2012) 9–22. [15] J. Chen, N. Imanaka, J. Chen, J.D. Griffin, Hypoxia potentiates Notch signaling in breast cancer leading to decreased E-cadherin expression and increased cell migration and invasion, Br. J. Cancer 102 (2010) 351–360. [16] P. Baskaran, C.Y. Huang, V.V. Padma, Neferine prevents NF-κB translocation and protects muscle cells from oxidative stress and apoptosis induced by hypoxia, Biofactors 42 (2016) 407–417. [17] V.C. Wong, J.L. Morse, A. Zhitkovich, p53 activation by Ni (II) is a HIF-1α independent response causing caspases 9/3-mediated apoptosis in human lung cells, Toxicol. Appl. Pharmacol. 269 (2013) 233–239. [18] M. Jawahir, S.A. Nicholas, Coughlan, V.V. Sumbayev, Apoptosis signal-regulating kinase 1 (ASK1) and HIF-1α protein are essential factors for nitric oxide-dependent accumulation of p53 in THP-1 human myeloid macrophages, Apoptosis 13 (2008) 1410–1416. [19] N. Goda, M. Kanai, Hypoxia-inducible factors and their roles in energy metabolism, Int. J. Hematol. 95 (2012) 457–463. [20] , Y. Kim, Yi, , , Curcumin Induces Apoptosis by the Regulation of AMPKα1 and HIF-1α Levels Journal of Cancer Prevention 17 2012 116–120. [21] X. Zhang, J.H. Cui, W. Zhou, S.S. Li, Design, synthesis and anticancer activity of

Fig. 9. Molecular docking analysis of PMMB232, showing proposed binding modes at the HIF-1α (PDB code: 4W9I). (A) Interaction of PMMB232 with the amino acid residues (carbon atom, gray; oxygen atom, red; hydrogen atom, white; chlorine atom, green and nitrogen atom; light blue) (B) Binding position of PMMB232 in the protein surface of HIF1α (carbon atom, gray; oxygen atom, red; hydrogen atom, white; chlorine atom, green and nitrogen atom; light blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

code: 4W9I) were performed. In the view of interaction energy, PMMB232 had a best binding free energy of −48.856 kcal/mol. It meant that the binding of PMMB232 with HIF-1α was more stable than those of others. All these results demonstrated that PMMB232 could be developed as a potential anticancer agent.

5. Conclusion PMMB232, one of series of shikonin derivatives by exerting inhibitory effects on apoptosis, oxidative stress, mitochondrial potential, adhesive ability, wound-healing migration, expression of HIF-1α and glycolysis-related proteins, exhibits therapeutic potential in anticancer experiments. Our findings suggest that shikonin derivatives could induce apoptosis in cancer cells through the mechanisms of action of HIF1α. The present study points out a new target for research of shikonin derivatives in anti-cancer capacity.

Authors' contributions HWH, GHL, XMW and YHY conceived and designed the study. HWH, WXS and SJC collected the samples. HWH, ZSZ and LJH performed the experiments. HWH, JLQ and RWY analyzed the data. HWH and ZSZ wrote the paper. GHL, XMW and YHY revised the paper. All authors read and approved the final manuscript.

665

Biomedicine & Pharmacotherapy 97 (2018) 656–666

H.-W. Han et al.

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36] M.X. Ye, Y.L. Zhao, Y. Li, Q. Miao, Z.K. Li, X.L. Ren, L.Q. Song, H. Yin, J. Zhang, Curcumin reverses cis-platin resistance and promotes human lung adenocarcinoma A549/DDP cell apoptosis through HIF-1 alpha and caspase-3 mechanisms, Phytomedicine 19 (2012) 779–787. [37] C. Tan, L. Zhang, X. Cheng, X.F. Lin, R.R. Lu, J.D. Bao, H.X. Yu, Curcumin inhibits hypoxia-induced migration in K1 papillary thyroid cancer cells, Exp. Biol. Med. 240 (2015) 925–935. [38] M.X. Ye, Y.L. Zhao, Y. Li, Q. Miao, Z.K. Li, X.L. Ren, L.Q. Song, H. Yin, J. Zhang, Curcumin reverses cis-platin resistance and promotes human lung adenocarcinoma A549/DDP cell apoptosis through HIF-1 and caspase-3 mechanisms, Phytomedicine 19 (2012) 779–787. [39] H.Y. Lin, H.W. Han, L.F. Bai, H.Y. Qiu, D.Z. Yin, J.L. Qi, X.M. Wang, H.W. Gu, Y.H. Yang, Design, synthesis and biological evaluation of shikonin thio-glycoside derivatives: new antitubulin agents, RSC Adv. 4 (2014) 49796–49805. [40] G.L. Semenza, Hypoxia-inducible factor 1: Regulator of mitochondrial metabolism and mediator of ischemic preconditioning, BBA Mol. Cell Res. 7 (2011) 1263–1268. [41] K.J. Morten, L. Badder, H.J. Knowles, Differential regulation of HIF-mediated pathways increases mitochondrial metabolism and ATP production in hypoxic osteoclasts, J. Pathol. 229 (2013) 755–764. [42] K. Chu, K.M. Boley, R. Moraes, S.H. Barsky, F.M. Robertson, The Paradox of ECadherin Role in response to hypoxia in the tumor microenvironment and regulation of energy metabolism, Oncotarget 4 (2013) 446–462. [43] D.K. Afriyie, G.A. Asare, K. Bugyei, J. Lin, J. Peng, Z. Hong, Mitochondria-dependent apoptogenic activity of the aqueous root extract of Croton membranaceus against human BPH-1 cells, Genet. Mol. Res. 14 (2015) 149–162. [44] D.D. Nguyen, G. Kim, E.K. Pae, Modulation of muscle fiber compositions in response to hypoxia via pyruvate dehydrogenase kinase-1, Front. Physiol. 7 (2016) 1–11. [45] E.L. Bell, B.M. Emerling, S.J.H. Ricoult, L.G. Uarente, SirT3 suppresses hypoxia inducible factor 1 alpha and tumor growth by inhibiting mitochondrial ROS production, Oncogene 26 (2011) 2986–2996. [46] F. Yang, H.F. Zhang, Y.D. Mei, M. Wu, Reciprocal regulation of HIF-1 alpha and lincRNA-p21 modulates the warburg effect, Mol. Cell 1 (2014) 88–100. [47] X. Liang, X.Q. Xu, F.C. Wang, N. Li, J.M. He, E-cadherin increasing multidrug resistance protein 1 via hypoxia-inducible factor-1α contributes to multicellular resistance in colorectal cancer, Tumor Biol. 37 (2016) 425–435. [48] P. Maroni, E. Matteucci, L. Drago, G. Banfi, P. Bendinelli, M.A. Desiderio, Hypoxia induced E-cadherin involving regulators of Hippo pathway due to HIF-1α stabilization/nuclear translocation in bone metastasis from breast carcinoma, Exp. Cell Res. 330 (2015) 287–299. [49] J. Yang, X. Zhang, Y. Zhang, D.M. Zhu, Y. L. F. Zhang, Y.B. Li, D.C. Zhu, J. Li, HIF-2α promotes epithelial-mesenchymal transition through regulating Twist2 binding to the promoter of E-cadherin in pancreatic cancer, J. Exp. Clin. Cancer Res. 35 (2016) 1–10. [50] J.C. Cheng, C. Klausen, P.C.K. Leung, Hypoxia-inducible factor 1 alpha mediates epidermal growth factor-induced down-regulation of E-cadherin expression and cell invasion in human ovarian cancer cells, Cancer Lett. 329 (2013) 197–206.

shikonin and alkannin derivatives with different substituents on the naphthazarin scaffold, Chem. Res. Chin. Univ. 31 (2015) 394–400. L.J. Kuo, C.Y. Huang, W.L. Cheng, C.S. Hung, C.T. Wu, F.Y. Lin, Y.J. Chang, M.T. Huang, Glucose-regulated protein 78 mediates the anticancer efficacy of shikonin in hormone-refractory prostate cancer cells, Tumor Biol. 36 (2015) 5063–5070. T.J. Lin, H.T. Lin, W.T. Chang, P. Mitapalli, P.W. Hsiao, S.Y. Yin, N.-S. Yang, Shikonin-enhanced cell immunogenicity of tumor vaccine is mediated by the differential effects of DAMP components, Mol. Cancer 14 (2015) 174–187. H.Y. Lin, Z.K. Li, L.-F. Bai, S.K. Baloch, F. Wang, H.Y. Qiu, X. Wang, J.L. Qi, R.W. Yang, X.M. Wang, Y.H. Yang, Synthesis of aryl dihydrothiazol acyl shikonin ester derivatives as anticancer agents through microtubule stabilization, Biochem. Pharmacol. 96 (2015) 93–106. J. Guo, X.F. Chen, J. Liu, H.Y. Lin, H.W. Han, H.C. Liu, S.C. Huang, B.K. Shahla, A. Kulek, J.L. Qi, X.M. Wang, L.J. Ling, Y.H. Yang, Novel shikonin derivatives targeting tubulin as anticancer agents, Chem. Biol. Drug Des. 84 (2014) 603–615. S.K. Baloch, L.J. Ling, H.Y. Qiu, L. Ma, H.Y. Lin, S.C. Huang, J.L. Qi, X.M. Wang, G.H. Lu, Y.H. Yang, Synthesis and biological evaluation of novel shikonin ester derivatives as potential anti-cancer agents, RSC Adv. 4 (2014) 35588–35596. R. Tian, Y. Li, M. Gao, Shikonin causes cell-cycle arrest and induces apoptosis by regulating the EGFR–NF-κB signalling pathway in human epidermoid carcinoma A431 cells, Biosci. Rep. 35 (2015) 1–13. D.Z. Duan, B.X. Zhang, J. Yao, Y.P. Liu, J.G. Fang, Shikonin targets cytosolic thioredoxin reductase to induce ROS-mediated apoptosis in human promyelocytic leukemia HL-60 cells, Free Radic. Biol. Med. 70 (2014) 182–193. D. Hara, H. Komatsu, A. Son, S.I. Nishimoto, K. Tanabe, Water-Soluble phosphorescent ruthenium complex with a fluorescent coumarin unit for ratiometric sensing of oxygen levels in living cells, Bioconjugate Chem. 26 (2015) 645–649. X.M. Peng, G.L.V. Damu, C.H. Zhou, Current developments of coumarin compounds in medicinal chemistry, Curr. Pharm. Design 19 (2013) 3884–3930. L. Pisani, R. Farina, M. Catto, R.M. Iacobazzi, O. Nicolotti, S. Cellamare, G.F. Mangiatordi, N. Denora, R.S. Otero, L. Siragusa, C.D. Altomare, A. Carotti, Exploring basic tail modifications of coumarin-Based dual acetylcholinesteraseMonoamine oxidase B inhibitors: identification of water-Soluble brain-Permeant neuroprotective multitarget agents, J. Med. Chem. 59 (2016) 6791–6806. J. Dandriyal, R. Singla, M. Kumar, V. Jaitak, Recent developments of C-4 substituted coumarin derivatives as anticancer agents, Eur. J. Med. Chem. 119 (2016) 141–168. G.Y. Ciftci, E. Senkuytu, M. Bulut, M. Durmus, Novel coumarin substituted water soluble cyclophosphazenes as turn-Off type fluorescence chemosensors for detection of Fe3+ ions in aqueous media, J. Fluoresc. 25 (2015) 1819–1830. F.G. Medina, J.G. Marrero, M.M. Alonso, M.C. Gonzalez, I.C. Guerrero, A.G.T. Garcia, S.O. Robles, Coumarin heterocyclic derivatives: chemical synthesis and biological activity, Nat. Prod. Rep. 32 (2015) 1472–1507. B.Y. Wang, X.Y. Liu, Y.L. Hu, Z.X. Su, Synthesis and photophysical behavior of a water-soluble coumarin-bearing polymer for proton and Ni2+ ion sensing, Polym. Int. 58 (2009) 703–709.

666