Dehydrocostus lactone inhibits in vitro gastrinoma cancer cell growth through apoptosis induction, sub-G1 cell cycle arrest, DNA damage and loss of mitochondrial membrane potential

Abstract
Introduction: The purpose of the present study was to evaluate the antipro- liferative activity of dehydrocostus lactone against human BON-1 cancer cell lines and to explore the possible underlying mechanism. Material and methods: MTT cell viability assay was used to determine cy- totoxic effects of dehydrocostus lactone in BON-1 cells. Fluorescence and transmission electron microscopic (TEM) techniques were used to study the effect of the compound on cellular morphology and apoptosis. Flow cytom- etry was used to assess the effect on cell cycle phase distribution. Effects of the drug on cell apoptosis and mitochondrial membrane potential were analyzed by flow cytometry using annexin v and rhodamine-123 as fluores- cent probes.
Results: The results of the present study indicated that dehydrocostus lac- tone significantly (p < 0.01) inhibited the growth of BON-1 cancer cells. These growth inhibitory effects of dehydrocostus lactone on BON-1 were found to be time and concentration-dependent. The IC50 of dehydrocostus lactone were found to be 71.9 μM and 52.3 μM at 24 and 48 h time intervals respectively. The growth inhibitory effects of dehydrocostus lactone were found to be due to loss of mitochondrial membrane potential, the induction of apoptosis and sub-G1 cell cycle arrest. Conclusions: Dehydrocostus inhibits in vitro gastrinoma cancer cell growth and therefore may prove beneficial in the management of gastrinoma cancer.

Introduction
Gastrinoma arises as a result of a tumor in the duodenum or pancreas which secretes surplus amounts of gastrin, leading to ulcers in the stom- ach or duodenum. Gastrinoma mostly originates in the pancreas but can also be found in the duodenum or stomach. Pancreatic gastrinoma is characterized by a higher tendency to be malignant. Gastrin is a hormone that controls the amount of acid in the stomach. Gastrinomas produce enormous amounts of gas- trin, and this causes the stomach to make more acid. Gastrinomas are a type of neuroendocrine tumors that develop in cells that are triggered by nerve cells to produce hormones. Gastrinomas are also known as pancreatic neuroendocrine tumors. Pancreatic neuroendocrine tumors (PNETs) are malignant tumors likely arising from islet cells of the pancreas [1, 2]. Pancreatic neuroendocrine tu- mors secrete a range of hormones including insu- lin, glucagon and somatostatin. However, several PNETs do not secrete any hormone [3]. Pancreatic neuroendocrine tumors can be very difficult to treat and can vary from benign to highly malig- nant. PNETs can be slow growing or aggressive in certain cases. Regarding PNET treatment, surgical resection alone is mostly useful for its cure in initial stag- es. However, in most PNET cases (about 50%) patients have an advanced stage of the disease and then suffer from uncontrolled hormone secre- tion giving rise to many other complications [4].

It has been reported that gastrinoma tumors show a good response towards the use of cytotoxic che- motherapeutic drugs especially the combination of cisplatin and etoposide [5, 6]. However, there are serious side-effects associated with these cy- totoxic anticancer drugs. Hence, there is a need to find alternative chemotherapeutic agents for the treatment of gastrinoma tumors. The objective of the present study was to evaluate the anticancer and apoptotic effects of dehydrocostus usual- ly isolated from Cichorium intybus in the BON-1 PNET cell line and to explore the possible underly- ing mechanism.Dehydrocostus lactone (> 98%) was purchased from Chengdu Preferred Biotech Co. Ltd (China) and was dissolved in dimethyl sulfoxide (DMSO) at the concentration of 50 mM, stored as small aliquots at –20°C. Different concentrations of the compound were prepared (0, 5, 25, 50, 75 and 100 μM) for cell culture experiments. 3-[4,5-di- meth-yl-2-thiazolyl]-2,5-diphenyl tetrazolium bro- mide (MTT) was purchased from Molecular Probes (USA). Dulbecco’s modified Eagle’s medium, fetal bovine serum (FBS), penicillin-streptomycin and rhodamine-123 (Rh-123) were obtained from Hangzhou Sijiqing Biological Products Co., Ltd, China. Propidium iodide (PI), trypsin, dimethylsulfoxide (DMSO) and Hoechst 33258 were pur- chased from Sigma-Aldrich (USA).The human pancreatic neuroendocrine tumor cell line (BON-1) was procured from the Cancer Research Institute of Beijing, China. The cells were grown in DMEM supplemented with 10% fetal calf serum (FBS) and 150 U/ml of penicillin. Incuba- tion of the cells was done at 37°C in a humidified atmosphere of 5% CO2 and 95% air.

The medium was stored at low temperature (2–5°C). The medi- um was replaced every 2 days. Cells were subcul- tured every 4 days.The cytotoxic effects of dehydrocostus lactone on BON-1 cell proliferation were determined by MTT assays. BON-1 cells (2 × 105 cells/well) were seeded and cultured with varying doses of dehy- drocostus lactone (0, 5, 25, 75, and 100 μM each) for 24 and 48 h. Following drug treatment, the medium was changed and 3-(4,5-dimethylthi- azol-zyl)-2,5-diphenyltetrazolium bromide (MTT: 2 mg/ml) was added for 3 h. The number of viable cells is equal to the formation of formazan crystals which were dissolved in ethanol and the optical density was measured on a microplate reader (ELX 800; Bio-tek Instruments, Inc., Winooski, VT, USA) at a wavelength of 490 nm. The effects of dehy- drocostus lactone on cell viability were calculat- ed as an inhibition ratio (I %) using the following equation (optical density at 490 nm):I% = [OD490 (Control) – OD490 (Treated)] × 100% [OD490 (Control)Detection of apoptosis using fluorescence microscopyBON-1 cells (2 × 105 cells/well) were plated in six-well plates and then cultured for 24 h to al- low complete attachment to the surface of the plates. The cells were treated with various doses of dehydrocostus lactone treatment (0, 5, 50 and 100 μM) for 48 h and then stained with Hoechst 33258 (2 μg/ml) at 37°C for 20 min.

Nuclear mor- phology was examined under a fluorescence mi- croscope (Olympus, Tokyo, Japan) to identify cells undergoing apoptosis.BON-1 cells were treated with or without dehy- drocostus lactone for 48 h and fixed in sodium cac- odylate (pH 7.4) and glutaraldehyde solution for 3 h. The cells were washed in PBS, then fixed in1.5% OsO4 solution for 1 h at 25 C, washed and then dehydrated in an ethanol solution of increas- ing polarity (50% to 80% with 15 min of each bath). The cells were then embedded in EMbed 812 resin (SPI Supplies, PA, USA). After staining with ura- nyl acetate and lead citrate for 30 min, cell ultra- structure was analyzed (at 300 kV voltage) using ultra-thin sections with a transmission electron microscope (Hitachi High Technologies America, North 28th Avenue, Texas, United States).BON-1 cells were seeded in a 100-mm cell cul- ture dish for 48 h and treated with 0, 5, 50 and 100 μM dehydrocostus lactone for 48 h. The cells were harvested and washed with PBS, and the pellets were lysed with a 200 μl DNA lysis buffer (20 mM EDTA, 40 mM Tris-HCl) for 20 min. After centrifugation, the supernatants were prepared in an equal volume of 3% sodium-dodecyl sulfate, incubated with 3 mg/ml RNase A at 55°C for 3 h followed by digestion with 2.5 mg/ml proteinase K for 2 h at 20oC. Subsequent to the addition of 10 M ammonium acetate, the DNA was precipi- tated with cold ethanol and collected by centrif- ugation at 15,000 × g for 15 min.

DNA was then dissolved in gel loading buffer, separated by elec- trophoresis in 1.5% agarose gel and visualized un- der UV light, following ethidium bromide staining.Mitochondrial membrane potential (MMP) in the human pancreatic neuroendocrine tumor cell line (BON-1) was measured with rhodamine-123 dye, which preferentially enters the active mito- chondria based on the highly negative MMP. The cells were seeded in 96-well plates at a density of 1 × 105 cells/ml. The cells were treated with vary- ing doses of dehydrocostus lactone (0, 5, 50 and 100 μM). Depolarization of MMP results in the loss of rhodamine 123 from mitochondria and a de- crease in intracellular fluorescence is observed. Rhodamine 123 (final concentration of 10 μM) was added to the harvested cells and analyzed us- ing a FACSCalibur instrument (BD Biosciences, San Jose, CA, USA) equipped with Cell Quest 3.3 soft- ware.twice with PBS. After that the cells were fixed with 70% cold ethanol overnight and then treated with 20 μg/ml RNase A, then stained with 3 μg/ml of propidium iodide. Finally the DNA content and cell cycle distribution were analyzed by flow cytome- try. The experiments were repeated three times. The cell cycle analysis was performed by a FACS- Calibur instrument (BD Biosciences, San Jose, CA, USA, equipped with Cell Quest 3.3 software with DNA propidium iodide (PI) staining.Data are presented as the mean ± SEM of the control. All experiments were repeated at least three times. The differences between groups were analyzed by one-way ANOVA with Tukey’s post hoc tests. Significance of difference was indicated as *p < 0.05, **p < 0.01.

Results
Antiproliferative effect of dehydrocostus lactone on the cytotoxicity of human pancreatic neuroendocrine tumor cell lines (BON-1)Figure 1 shows that the dehydrocostus lactone (Figure 2) exhibited dose-dependent as well as time-dependent growth inhibitory effects in these cells. The efficacy of the compound was evaluated by determining its IC50 value, which was found to be 71.9 μM and 52.3 μM at 24 and 48 h time in- tervals respectively.After treating BON-1 cells with 0, 5, 50 and 100 μM doses of the compound, the cells revealed apoptotic morphological features including nu- clear fragmentation and condensation, cellularThe effect of dehydrocostus lactone on cell cycle phase distribution was evaluated by flow cytometry using propidium iodide as a probe. BON-1 cells (2 × 105 cells/ml) were seeded in 60-mm dishes and treated with 0, 5, 50 and 100 μM of dehydrocostus lactone for 48 h. After treatment, the cells were trypsinized and washedapoptosis in BON-1 cancer cells was further con- firmed by using TEM. The results of this observa- tion are shown in Figures 4 A–D. Control cells (un- treated cells) exhibited normal ultrastructure with round nuclear membrane, normal mitochondria and the nucleolus present within the nucleus (Fig- ure 4 A). But, after the cells were treated with 5, 50 and 100 μM concentrations of dehydrocostus lac- tone for 48 h, the mitochondria were destroyed, chromatin condensation occurred and the nuclear membrane disappeared. This was accompaniedby the appearance of vacuoles in the cytoplasm, which is a hallmark of early apoptotic events which was clearly absent in untreated control cells (Figures 4 B–D).

Thus TEM results confirm the re- sults of fluorescence microscopy, which indicates that dehydrocostus lactone induces apoptosis in BON-1 cancer cells.DNA fragmentation analysis induced by differ- ent doses of dehydrocostus lactone was observed by the formation of a DNA ladder using 1.5% aga- rose gel electrophoresis. DNA ladder formation was consistent with increasing doses of the com- pound, but no such DNA laddering was visible in the control group (Figure 5). However, increasing doses of the compound for 48 h led to a signif- icant increase in DNA fragmentation. The DNA fragmentation is a sign of the apoptotic process which starts within the cell, further confirmingthat the dehydrocostus lactone induced cell death via apoptosis.Loss of mitochondrial transmembrane poten- tial is a crucial step in the intrinsic apoptotic event. In this study, we evaluated the effect of dehydro- costus lactone on the mitochondrial membrane potential loss in BON-1 cells using a fluorescent probe, Rh-123 and flow cytometric analysis. After the drug treatment, a significant increase in cells with decreased membrane potential was seen af- ter 48 h (Figures 6 A–D). The fraction of cells with decreased m seemed to follow the concen- tration of the compound. The percentage of cells with decreased m increased from 3.58% in the control group to 14.2%, 24.6% and 43.3% in 5, 50 and 100 μM-dehydrocostus lactone treated cells respectively. In cells, loss of m leads to mem-brane rupture followed by cytochrome c release and pro-apoptotic factors.Dehydrocostus lactone induces sub-G1 cell cycle arrest in BON-1 cellsFigure 7 indicates that the compound increased the percentage of sub-G1 cells from 8.1% in the untreated group to 14.6%, 28.5% and 56.9% in groups treated with 5, 50 and 100 μM doses of de- hydrocostus lactone respectively. The percentage of sub-G1 cells is an indication of the apoptotic cells which arise as a result of the treatment with the drug.

Discussion
Gastrinoma is a very lethal type of cancer and is very difficult treat like, many other types of cancers. Moreover, treatment options for gastrino- ma are limited and are often associated with nu- merous side effects [1]. The findings of this study indicate that dehydrocostus lactone is a potent cytotoxic agent which induces significant growth inhibitory effects in human pancreatic neuroendo- crine tumor cell lines (BON-1) in a time-dependent as well as dose-dependent manner. Our studies are also consistent with previous studies where- in sesquiterpene lactones have been reported to exhibit anticancer activities against a range of cancer types [7]. It has also been reported that these lactones exhibit selective cytotoxic activity against cancer cells and exhibit high IC50 values for normal cell lines indicating lower cytotoxicity for normal cells [8]. Moreover, to evaluate the effect of the solvent, we determined the cytotoxic ef- fects of DMSO on BON-1 and the results indicated that DMSO had negligible antiproliferative effects (Figure 8). Fluorescence microscopy using Hoechst 33258 showed that the compound induced apop- totic morphological features in these cells char- acterized by nuclear fragmentation and conden- sation, cellular shrinkage, membrane blebbing, etc. Further, transmission electron microscopy indicated that dehydrocostus lactone treatment at increasing doses led to formation of vacuoles, disappearance of nuclear membrane and dam- aged mitochondria. The effect intensified with increased doses of the compound.

The untreated cells, however, showed normal cellular morpholo- gy with round nuclei and intact mitochondria. Fur ther, using 1.5% agarose gel electrophoresis, DNA fragmentation analysis induced by dehydrocostus lactone was carried out. The results of this experi- ment indicated that as compared to the untreated control which did not show any signs of DNA frag- mentation, dehydrocostus lactone treated cells at increasing doses indicated significant levels of DNA fragmentation. Consistent with this, several studies have reported induction of apoptosis by dehydrocostus lactone and costunolide in many types of cancer cells [7–9].Dehydrocostus lactone also induced significant loss of mitochondrial membrane potential (m loss) in these cells. The percentage of cells with decreased m seemed to follow the compound dose. The percentage of cells with decreasedm increased from 3.58% in the control group to 14.2%, 24.6% and 43.3% in 5, 50 and 100 μM-de-hydrocostus lactone treated cells respectively. In cells, loss of m leads to membrane rupture fol- lowed by cytochrome c release and pro-apoptot- ic factors. Loss of mitochondrial transmembrane potential is a crucial step in the intrinsic apoptotic event [10, 11]. T

his compound also led to sub-G1cell cycle arrest in BON-1 cells, the percentage of sub-G1 cells increasing from 8.1% in the untreated group to 14.6%, 28.5% and 56.9% in groups treat- ed with 5, 50 and 100 μM doses of dehydrocostus lactone respectively.It has been reported that the process of cancer development is closely linked to the biochemical processes of cell cycle and apoptosis. Currently, the drug discovery process focuses on chemother- apeutic agents which can target the cell cycle and apoptosis [9–14].Dehydrocostus lactone belongs to the sesqui- terpene lactone class of naturally occurring com- pounds. They are sesquiterpenoids and contain a lactone ring. Sesquiterpene lactones are found in several plants and are well known for causing aller- gic reactions [15, 16]. Dehydrocostus lactone along with costunolide has been reported to induce apop- tosis and cell cycle arrest in human soft tissue sar- coma cells. Dehydrocostus lactone caused sub-G1 cell cycle arrest, as evident from the increased num- ber of sub-G1 cells in the treated cells. A previousstudy reported that a similar type of compound causes G2/M cell cycle arrest [17, 18]. Moreover, the current study used a different and rare cancer cell line, viz. BON-1 (human pancreatic neuroendo- crine tumor cell lines), which may be one reason for the cell cycle results reported in the present study. The findings of the present study report a different UNC0379 mechanism of action of dehydrocostus lactone. It induced sub-G1 cell cycle arrest and also induced loss of mitochondrial membrane potential, indicat- ing the intrinsic apoptotic pathway.

In conclusion, the findings of the present study reveal that dehydrocostus lactone inhibited can- cer cell growth in human pancreatic neuroendo- crine tumor cell lines via the induction of mito- chondrial mediated apoptosis, cell cycle arrest, DNA fragmentation and loss of mitochondrial membrane potential.