Selective cytotoxicity of epidithiodiketopiperazine DC1149B, produced by marine-derived Trichoderma lixii on the cancer cells adapted to glucose starvation

The core of solid tumors is characterized by hypoxia and a nutrient-starved microenvironment and has gained much attention as targets of anti-cancer drugs. In the course of search for selective growth inhibitors against the cancer cells adapted to nutrient starvation, epidithiodiketopiperazine DC1149B (1) together with structurally related compounds, trichodermamide A (2) and aspergillazine A (3), were isolated from culture extract of marine-derived Trichoderma lixii. Compounds 1 exhibited potent selective cytotoxic activity against human pancreatic carcinoma PANC-1 cells cultured under glucose-starved conditions with IC50 values of 0.02 µM. The selective index of the compound 1 was found to be 35,500-fold higher for cells cultured under glucose-starved conditions than those under the general culture conditions. The mechanistic analysis indicated that compound 1 inhibited the response of the ER stress signaling. In addition, these effects of compound 1 could be mediated by inhibiting complex II in the mitochondrial electron transport chain.


Introduction
Solid tumors contain hypoxic and nutrient-starved regions due to the abnormal cell proliferation coupled with the defective and disorganized vascular supply [1]. The cancer cells that have adapted to this tumor microenvironment are assumed to develop aggressive phenotype, impaired angiogenesis, and drug resistance [2,3]. Since the hypoxic and nutrient-starved tumor microenvironment differs significantly from the normal tissues, the compounds with selective cytotoxicity against the cancer cells in tumor microenvironment would have great therapeutic potential.
The marine flora and fauna are a rich source of therapeutic drugs because of their chemical and biological diversity. Some of the marine-derived compounds have been reported to inhibit the growth of cancer cells adapted to the hypoxic or nutrient-starved conditions. For example, furospinosulin-1 (furanosesterterpene) and dictyoceratins-A and -C (sesquiterpene phenols) isolated from the marine sponge Dactylospongia elegans were shown to exhibit the selective growth inhibitory activities against the hypoxia-adapted human prostate cancer cell line DU145 [4,5]. Recently, we reported the cytotoxic activities of polybrominated diphenyl ethers, N-methylniphatin A (new 3-alkyl pyridine alkaloid) and biakamides (unique new polyketides) isolated from Indonesian marine sponges of Dysidea sp., Xestospongia sp., and Petrosaspongia sp., respectively, against human pancreatic carcinoma cell line PANC-1 adapted to glucosedeficient growth conditions [6][7][8]. In this study, we present the isolation of epidithiodiketopiperazine DC1149B (1), and structurally related but inactive compounds, trichodermamide A (2) and aspergillazine A (3) isolated from the culture extract of marine-derived Trichoderma lixii., and the cytotoxic activity of compound 1 on nutrient-starved cancer cells, and propose the plausible mechanism of its action.

Cell culture and bioassay
Human pancreatic carcinoma cell line PANC-1 was maintained in the DMEM supplemented with heat-inactivated 10% FBS and kanamycin (50 µg/ml) in a humidified atmosphere with 5% CO 2  The bioassay was carried out according to a method described previously [6]. Briefly, PANC-1 cells (1 × 10 4 cells/100 µl in 96 well plastic plate) were pre-incubated in the DMEM supplemented with 10% FBS for 24 h. The medium was then replaced with either the General Glucose Medium or Glucose Deficient Medium to induce cells adaption to the nutrient starvation. After 12-h incubation, the serially diluted samples were added, and the cells were incubated for an additional 12 h in a humidified atmosphere with 5% CO 2 at 37 °C. The cell proliferation was detected using the WST-8 colorimetric reagent. The IC 50 value was determined by linear interpolation from the growth inhibition curve. We assessed the selectivity of the cytotoxic Western blotting analysis PANC-1 cells (3 × 10 5 cells/2 ml in 6-well plastic plate) were pre-incubated in the DMEM supplement with 10% FBS for 24 h. The medium was then replaced with either General Glucose or Glucose Deficient Medium. After 12-h incubation, compound 1 (0.02-0.3 µM) or antimycin A (0.3 nM, positive control) was added to each well and the cells were incubated for an additional 12 h in a humidified atmosphere with 5% CO 2 at 37 °C. Subsequently, the cells were rinsed with ice-cold PBS and lysed in the lysis buffer [50 mM Tris-HCl (pH 8.0) containing 150 mM NaCl, 5 mM EDTA, 1% glycerol, 1% NP-40, 1% protease inhibitor cocktail, and 1% phosphatase inhibitor cocktail]. The cell lysate was analyzed by SDS-PAGE and transferred onto PVDF membranes (GE Healthcare Life Sciences Buckinghamshire, UK). The membranes were then incubated with appropriate primary antibodies followed by HRP-conjugated secondary antibodies, and the immunopositive bands were visualized using an ECL kit (GE Healthcare Life Sciences). The luminescent signals were analyzed using an Image Quant LAS4010 Scanner (GE Healthcare Life Sciences).

Effects of compound 1 on the oxygen consumption of PANC-1 cells
Oxygen consumption rate of PANC-1 cells was assayed using OCR Assay Kit (Cayman Chemical, Ann Arbor, MI, USA) as per the manufacturer's instructions. Briefly, precultured PANC-1 cells (8.0 × 10 4 cells) in the black, clear bottom 96-well plate (Corning Incorporated, NY, USA) were incubated with General Glucose Medium for 18 h at 37 °C. The cells were replenished with 140 µl of fresh medium, and the test compound was added. This was followed by adding the phosphorescent probe to measure the oxygen consumption. 100 µl of mineral oil was added in each well to restrict oxygen diffusion. The signals were measured by an Infinite M1000 microplate reader (Tecan Group Ltd., Mannedorf, Switzerland) using time-resolved mode at Ex 380 nm and Em 650 nm for 180 min with 1-min time interval. The linear regression was applied after subtracting the blank, and the oxygen consumption rate was indicated by the slope of each signal profile.

Statistical analysis
Data are shown as means ± standard errors of n = 3 independent experiments, and the differences between data sets were assessed by Dunnett's test. Differences with p values of less than 0.05 were considered significant.

Results and discussion
Cytotoxic activity of compounds 1-3 against the PANC-1 cells cultured under both glucose-deficient conditions and general culture conditions The bioassay-guided separation of the active 90% MeOHsoluble portion from culture extract of marine-derived Trichoderma lixii 15G49-1 led to the isolation of DC1149B (1), trichodermamide A (2), and aspergillazine A (3) (Fig. 1). We then evaluated the cytotoxic activities of compounds 1-3 against the PANC-1 cells cultured under both glucose-deficient and general culture conditions. Antimycin A, which inhibits the growth of PANC-1 cells adapted to the nutrient-starved conditions, was used as the positive control [14].
As shown in Table 1, DC1149B (1) showed a higher cytotoxic activity against the PANC-1 cells adapted to glucose starvation (IC 50 = 0.02 µM) when compared with cells cultured with General Glucose Medium (IC 50 = 710 µM). The S. I. value of compound 1 was evaluated to be more than 35,000. On the other hand, the cytotoxic activities of trichodermamide A (2) and aspergillazine A (3) were very weak under glucose-starved conditions. These results indicated that compounds 1 had selective cytotoxic activity against the PANC-1 cells adapted to nutrient starvation. DC1149B (1) was first reported as a secondary metabolite of Trichoderma longibrachiatum and showed weak cytotoxic activity against rat adrenal pheochromocytoma cell line PC12 (IC 50 = 21.9 µg/ml, 42.4 µM) and human cervix epithelioid carcinoma HeLa cells (IC 50 = 50.1 µg/ml, 97.1 µM) [9]. H. Yamazaki et al. also showed that compound 1 exhibited moderate cytotoxicity on the human leukemic Jurkat cell line with IC 50 value of 5.1 µM [10]. For the first time, DC1149B (1) was shown to exhibit the preferential cytotoxic activity against PANC-1 cells adapted to glucose starvation. We next explored the underlying mechanism of DC1149B (1) cytotoxicity.

Effects of compound 1 on the Akt signaling and induction of GRP78
Recent studies on cancer cells adapted to nutrient starvation have revealed that the activation of PI3k/Akt/mTOR signaling pathway and the unfolded protein response (UPR) such as induction of glucose-regulated protein 78 (GRP78) were important for the adaptation of cancer cells to nutrient starvation [15,16]. Therefore, these processes have attracted much attention as possible drug targets for cancer chemotherapy. This observation prompted us to investigate the effect of compound 1 on the Akt signaling and the induction of GRP78 by utilizing a western blotting method (Fig. 2). Antimycin A was used as a positive control for comparison. The PANC-1 cells cultured in the Glucose Deficient Medium increased the expression levels of phosphorylated Akt and GRP78 proteins compared with cells cultured in the General Glucose Medium (Fig. 2, lanes 1 and 2). Treatment with DC1149B (1) at the lowest test concentration of 0.02 µM decreased the expression of GRP78 in the nutrient-starved cells to the basal level of cells cultured in General Glucose Medium (Fig. 2, lanes 5-10). Also, treatment with DC1149B (1) reduced the levels of Akt and phosphorylated Akt in the nutrient-starved cells in a dose-dependent manner (Fig. 2,  lanes 5-10). These results indicated that DC1149B (1) inhibited response of ER stress signaling. Besides, compound 1  (Fig. 2, lanes  3 and 4). The present results, therefore, suggest that compound 1 showed a similar behavior with that of antimycin A against the PANC-1 cells cultured in the Glucose Deficient Medium.

Effect of compound 1 on the mitochondrial electron transport chain
Antimycin A is a known inhibitor of complex III in the mitochondrial electron transport chain [17]. Therefore, we next examined the effect of compound 1 on the function of mitochondrial complex I-V using the Mito Check Complex Activity Assay Kit. As shown in Table 2, compound 1 inhibited complex II with IC 50 values of 19 µM, while other complexes remained unaffected up to 30 µM. It is known that inhibitors of the mitochondrial electron transport chain affect the oxygen consumption of the treated cells [18]. Since the inhibitory effect of compound 1 on the complex II was weak compared with that on the PANC-1 cells adapted to glucosestarved conditions (Table 1), we explored whether DC1149B (1) inhibited the oxygen consumption on the PANC-1 cells (Fig. 3). As a result, positive control of Antimycin A inhibited the oxygen consumption on the PANC-1 cells cultured in the General Glucose Medium, while the carbonyl cyanide m-chlorophenylhydrazone (CCCP, negative control), which is uncoupling substance of mitochondria, showed the opposite effect to antimycin A [18]. Compound 1 also inhibited the oxygen consumption on the PANC-1 cells under general culture conditions (Fig. 3). This result indicated that compound 1 inhibits complex II in the mitochondrial electron transport chain. We also evaluated the effect of compound 1 on the oxygen consumption of PANC-1 cells cultured in the Glucose Deficient Medium. However, the oxygen consumption rate of PANC-1 cells could not be determined, as the oxygen consumption of PANC-1 cells in the Glucose Deficient Medium was significantly lower than that in the General Glucose Medium. Collectively, our results indicate that the selective cytotoxicity of compound 1 might be mediated by the inhibition of complex II in the mitochondrial electron transport chain, while further study is necessary to consider whether complex II is major target for DC1149B (1) as a selective inhibitor against cancer cells adapted to glucose-starved conditions.

Conclusions
In our search for inhibitors against the cancer cells adapted to nutrient starvation, the epidithiodiketopiperazine DC1149B (1) was isolated from culture extract of marine-derived Trichoderma lixii. DC1149B (1) showed potent selective cytotoxicity against the cancer cells adapted to the glucosestarved conditions, with a S.I. value of 35,500. This study also explains the previously unknown mechanism of action of compound 1 on the cancer cells. Our results revealed that compound 1 inhibited the ER stress signaling, and that these effects of compound 1 could be mediated by the inhibition of complex II in the mitochondrial electron transport chain. Currently, research is underway to synthesize a probe to identify the major target of compound 1 in PANC-1 cells adapted to nutrient-starved conditions, and to observe the distribution of compound 1 in PANC-1 cells.  The medium was then replaced with each fresh medium, and the test compound was added followed by adding the phosphorescent probe to measure the oxygen consumption. The signals were measured by a Tecan infinite M1000 using time-resolved mode. Differences were considered significant at *p < 0.05 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.