Sensitization and synergistic anti-cancer effects of Furanodiene identified in zebrafish models

Furanodiene is a natural terpenoid isolated from Rhizoma Curcumae, a well-known Chinese medicinal herb that presents anticancer effects in various types of cancer cell lines. In this study, we have successfully established zebrafish xenografts with 5 various human cancer cell lines; and validated these models with anti-cancer drugs used clinically for treating human cancer patients. We found that Furanodiene was therapeutically effective for human JF 305 pancreatic cancer cells and MCF-7 breast cancer cells xenotranplanted into zebrafish. Furanodiene showed a markedly synergistic anti-cancer effect when used in combination with 5-FU (5-Fluorouracil) for both human breast cancer MDA-MB-231 cells and human liver cancer BEL-7402 cells xenotransplanted into zebrafish. Unexpectedly, Furanodiene reversed multiple drug resistance in the zebrafish xenotransplanted with cis-Platinum-resistant human non-small cell lung cancer cells and Adriamycin-resistant human breast cancer cells. Furanodiene played its anti-cancer effects through anti-angiogenesis and inducing ROS production, DNA strand breaks and apoptosis. Furanodiene suppresseed efflux transporter Pgp (P-glycoprotein) function and reduced Pgp protein level, but no effect on Pgp related gene (MDR1) expression. These results suggest sensitizition and synergistic anti-cancer effects of Furanodiene that is worthy of a further investigation.

Given the high genetic and physiological similarities with humans, zebrafish has been used as a powerful and cost-effective animal model for cancer research and cancer drug discovery [28][29][30][31][32][33] . Using zebrafish in cancer drug research and development provides several advantages including transparency, easy manipulation, high predictability, short testing period, and small amount of testing drugs required. Zebrafish cancer models could be established by carcinogen treatment, genetic knockout, gene overexpression or xenotransplanting human cancer cells into the zebrafish 33 . The translucent body of zebrafish enables researchers to visualize all processes associated with tumor formation, progression, metastasis and death [29][30][31][32][33] . The small molecule compounds or drugs can be added directly to the water environment of the zebrafish and the therapeutic effects could be assessed qualitatively and quantitatively.
In this study, we have successfully developed zebrafish xenografts with human A549 non-small-cell lung cancer cells, SGC-7901 stomach cancer cells, HepG2 liver cancer cells, JF 305 pancreatic cancer cells and MCF-7 breast cancer cells; and validated these models with anti-cancer drugs used clinically in treating cancer patients. We have also established zebrafish xenografts with 5 drug-resistant human gastric cancer cells. We found that the known anti-cancer agent Furanodiene isolated from traditional Chinese herbs are effective in treating human JF 305 pancreatic cancer cells and MCF-7 breast cancer cells xenotranplanted into the zebrafish. Furanodiene has clearly shown a synergistic anti-cancer effects when used in combination with 5-FU for both human breast cancer MDA-MB-231 cells and human liver cancer BEL-7402 cells xenotransplanted into the zebrafish. Surprisingly, we found that Furanodiene could reverse multiple drug resistance in zebrafish xenotransplanted with cis-Platinum-resistant human non-small cell lung cancer cells and Adriamycin-resistant human breast cancer cells. In the mechanistic studies, we have found that Furanodiene plays its anti-cancer roles through anti-angiogenesis and inducing ROS production, DNA strand breaks and apoptosis. We have also discovered that Furanodiene markedly suppresses efflux transporter P-glycoprotein (Pgp) function and reduces Pgp protein level, but no effect on Pgp related gene (MDR1) expression.

Results
Anti-cancer effects. To confirm whether response of zebrafish xenografts to cancer drugs is similar to the response of mammalian model systems, we treated the xenotransplanted zebrafish with each of 10 known anti-cancer drugs (Bevacizumab, cis-Platinum, Endostar, Paclitaxel, Vinorelbine, Adriamycin, 5-FU, Hydroxyurea, Irinotecan and Gemcitabine) and Furanodiene for 48 h. Three concentrations or dosages for each drug were assessed. The maximum tolerated concentration/dosages MTC/MTD of a testing drug was defined as a maximum concentration or maximum dosage that did not induce any observable adverse effect on zebrafish and was determined under a dissecting stereomicroscope by a well-trained zebrafish toxicologist. The MTC/ MTD was 400 ng for Bevacizumab, 2 ng for cis-Platinum, 80 ng for Endostar, 2.56 ng for Paclitaxel, 0.5 ng for Vinorelbine, 2.5 ng for Adriamycin, 2 ng for Irinotecan, 20 ng for Gemcitabine, 260 μg/mL for 5-FU and 1000 μg/ mL for Hydroxyurea, respectively and 12.2 ng for Furanodiene.
Xenotransplanted zebrafish were generated by microinjection of approximately 800 human cancer cells labeled with CM-Dil into the yolk sac of zebrafish that were at 48 hpf as we reported previously 34 . As expected, after a 48 h treatment, Bevacizumab, cis-Platinum, Endostar, Paclitaxel, Vinorelbine, Adriamycin, 5-FU, Hydroxyurea, Irinotecan and Gemcitabine all significantly inhibit tumor growth in the zebrafish xenotransplanted with various types of human cancer cells (Fig. 1). The cancer inhibition percentages in zebrafish xenotransplant model of A549 non-small-cell lung cancer cells were (4 ± 1.39)-(65 ± 2.54)% for Bevacizumab, ( were observed for all the tested anticancer drugs and Furanodiene (p < 0.05 or p < 0.01 or p < 0.001) (Fig. 2).
The mRNA levels of the Pgp related gene (MDR1) was determined in zebrafish treated with Furanodiene. The relative expression level of MDR1 gene was unchanged in zebrafish treated with Furanodiene at concentrations of 1/3 MTC and MTC as compared with control group, whereas the expression level of Pgp protein was severely decreased.

Dissussion
In this study, we have demonstrated that all the human anti-cancer drugs currently used clinically for cancer patients and selected for this study are therapeutically effective on the human cancer cells xenotranplanted into zebrafish, further validating and supporting zebrafish xenografts as a rapid, reliable, reproducible, and cost-efficient animal model for anti-cancer drug screening and assessment [34][35][36][37][38][39] . Our results from this investigation have confirmed that zebrafish xenotransplanted cancer models are also suitable for assessing sensitization and synergistic effects of anti-cancer drugs.
In the mechanistic studies, we have found that Furanodiene plays its anti-cancer role through anti-angiogenesis, inducing ROS production, DNA strand breaks and apoptosis. Neoangiogenesis has been recognized as a hallmark of tumor progression and about a dozen of anti-angiogenetic drugs have been globally marked [40][41][42][43][44] . Elevated ROS production has been indicated to damage large biological molecules including DNA, leading to DNA strand breaks and apoptosis [45][46][47] . Caspase 8 participates in the intracellular signaling cascade leading to apoptosis while caspase 9 has been linked to the mitochondrial death pathway [48][49][50][51] . Caspase 3/7 is either partially or totally responsible for the proteolytic cleavage of many key proteins during apoptosis and interacts with caspase 8 and caspase 9 [48][49][50][51] . Our results showed that the activity of caspases 8, 9 and 3/7 were all notablely increased in the cancer cell xenotransplanted zebrafish treated with Furanodiene, revealing that Furanodiene-induced zebrafish apoptosis are both caspase 8 and caspase 9 dependent, leading to cancer cell death.
Multiple-drug resistance is a most common and challengable problem in cancer treatment for both cytotoxic and targeted anti-cancer drugs 52 and there are no effective sensitizing drugs available in market 53 . Surprisingly, in this study, when cis-Platinum-resistant human non-small cell lung cancer cells and Adriamycin-resistant human breast cancer cells xenotransplanted into zebrafish are co-treated with Furanodiene, significant cancer www.nature.com/scientificreports www.nature.com/scientificreports/ suppression is demonstrated that are statistically more potent than either cis-Platinum/Adriamycin (no significant inhibition) or Furanodiene alone. We have also discovered that Furanodiene markedly suppresses efflux transporter Pgp function and reduces Pgp protein, but no effect on Pgp gene expression. These data suggest that Furanodiene could be a potential sensitizing agent to anti-cancer drugs probably mainly through blocking Pgp efflux transportation. These results are consistant with other reports indicating that Pgp is one of major players responsible for anti-cancer drug resistance and blocking Pgp could reverse the cancer drug resistence [54][55][56][57] , Further studies are under progresses to confirm Furanodiene pharmacology in mammalian models and to investigate other cellular, biochemical, and molecular mechanisms involved in Furanodiene anti-cancer pathways.

Materials and Methods
Zebrafish care and maintenance. Two lines of zebrafish were used in this study: wild-type AB line and Tg (fli1a: EGFP) y1 zebrafish. Adult zebrafish were housed and maintained in accordance with standard procedures. Zebrafish were generated by natural pair-wise mating according to the Zebrafish Handbook 34 . Four to five pairs of adult zebrafish were set up for each mating, and on average 200-300 zebrafish per pair were generated. Zebrafish were maintained at 28 °C in fish water (0.2% Instant Ocean Salt in deionized water, pH 6.9-7.2, conductivity 480-510 μS/cm, and hardness 53.7-71.6 mg/L CaCO 3 ). Zebrafish were cleaned and staged at 6 and 24 hours post fertilization (hpf) 34

Determination of maximum tolerated concentration/dosages (MTC/MTD). To determine MTC/
MTD of a testing drug, zebrafish were treated with a testing drug and mortality and toxicity were recorded at the end of treatment. In the initial tests, five concentrations (0.1, 1, 10, 100, and 500 μg/mL for soaking drugs and 0.1, 1, 10, 100, and 500 ng for injection drugs) were used for each drug. If a MTC/MTD could not be found from the initial tests, additional concentrations within the range of 0.01-2000 μg/mL or 0.01-2000 ng were tested. The MTC/MTD of a testing drug was defined as a maximum concentration or maximum dosage that did not induce any observable adverse effect on zebrafish and was determined under a dissecting stereomicroscope by a well-trained zebrafish toxicologist.
Cell culture. Human cancer cells were purchased from American Type Culture Collection (ATCC) and were subcultured and maintained in tissue culture flasks at 37 °C in a humidified, 5% CO2 atmosphere in RPMI 1640 essential medium or in DMEM essential medium (Gibco), supplemented with 10% heat-inactivated fetal bovine serum, 100 units/ml penicillin, 100 g/ml streptomycin, and 2 mM L-glutamine (Gibco) 34 .
Cell labeling. Human cancer cells were collected by centrifugation and resuspended in phosphate-buffered saline (PBS). The cells were fluorescently labeled by incubating with 10 µg/ml CM-DiI (Invitrogen, Burlington, ON, Canada) containing 0.5% DMSO and were adjusted to a density of 100 × 10 6 cells/ml (100 cells per nL) in HBSS. Labeled cells were injected within 2 hours. Cells prepared for injection in this way routinely had fewer than 5% dead cells and were in a single cell suspension, yielding controllable and reproducible numbers of injectable cells that were highly and uniformly fluorescent. The dye was transferred from mother to daughter cell and fluorescent single cells were clearly visible after several doublings on the periphery of late arising tumors 34 .
Cell transplantation. The transplantation protocol was similar to that described by our previous report 34 .
CM-DiI-labeled human cancer cells were loaded into a pulled glass micropipette (VWR blood capillaries #53508-400) that was drawn on an electrode puller and then trimmed to form a needle with a resulting internal diameter of approximately 15 micron and outer diameter of approximately 18 micron. The microneedle was attached to an air driven Cell Tram (Eppendorf). The tip of the needle was inserted into the yolk of a 48 hpf zebrafish and the pulse time controlled to deliver ~500 cells in 15 nL using positive pressure. The number of injected cells was standardized by fixing cell density and injection volume. After one hour recovery period at 28 °C, implanted zebrafish were examined under a fluorescence microscope (AZ100, Nikon, Japan) for the presence of xenotransplanted cells that reside only in the yolk and are then transferred to 35 °C for the duration of the experiment 34 . www.nature.com/scientificreports www.nature.com/scientificreports/ Determination of prolonged survival period. Xenotransplanted zebrafish were used to determine effects of drugs on the survival period of xenografted zebrafish. At 24 hpx xenotransplanted zebrafish were treated with a testing drug at 3 various concentrations (1/9 MTC, 1/3 MTC and MTC) for a treatment period of 9 days (d). The number of dead zebrafish in each group was recorded on a daily basis. Prolonged survival period was calculated based on the survival rate of xenotransplanted zebrafish treated with tested agents, as compared with untreated cancer model group. The prolonged survival period was calculated using the following formula: The prolonged survival period time = S(Drug)/S(Control).

Quanlification of synergistic anti-cancer effects.
To determine whether Furanodiene had a synergistic effect, zebrafish at 48 hpf was xenotransplanted with human liver cancer BEL-7402 or breast cancer MDA-MB-231 sensitive cancer cell lines labeled with CM-Dil. At 24 hpx, 5-Fluorouracil (5-FU) at 1/3 MTC, Furanodiene at 1/9 MTC, and Furanodiene at 1/9 MTC in combination with 5-FU at 1/3 MTC, respectively, were added to the treatment solution for a treatment period of 72 h. After treatment, xenotransplanted zebrafish were photographed under a fluorescence microscope and fluorescence intensity from the xenotransplanted cancer cells was quantified as described above 35,[37][38][39] . Reactive oxygen species (ROS) measurement. ROS levels in zebrafish treated with Furanodiene were analyzed using an oxidation sensitive probe, 5-(and 6-)-chloromethyl-20, 70-dichloro-dihydrofluoresceindiacetate (CM-H2DCFDA, Life Technologies, Carlsbad, CA). The treated zebrafish were incubated with 0.5 mg/mL CM-H2DCFDA for 1 h in dark at 28 °C. After rinsing for 3 times using fish water, zebrafish were transferred into a 96-well microplate (1 zebrafish per well) and ROS was measured at 488 nm under a multimode microplate reader (Berthold Technologies, Mithras LB940, Germany) as described by our group 39 .

Measurement of anti
DNA strand break analysis. After treatment, zebrafish were lyzed into a single cell suspension. Zebrafish cells were combined with agarose at 1:10 ratio (v/v), mixed well by pipetting and immediately transferred onto the slide. The slides were carefully transferred from either neutral solution (for double DNA strand break detection) or alkaline solution (for single DNA strand break detection) to a horizontal electrophoresis chamber and electrophoresed at 28 volts for 20 min. DNA Staining was done by Vista Green DNA Dye (Cell Biolabs, San Diego, USA) and the Olive tail moment (OTM) was utilized as a biomarker to quantify DNA strand as reported by one of our co-authors 58 .
Apoptosis quantifications. Acridine orange staining. AB wild type zebrafish at 6 hpf were distributed into 6-well microplates, 30 zebrafish per well in 3 mL fish water. Furanodiene at 3 various concentrations (1/9 MTC, 1/3 MTC and MTC) were added to the treatment solution for treatment period of 24 h. After treatment, zebrafish were stained with acridine orange and observed for apoptotic cells that would display yellow-green fluorescent spots under the fluorescence microscope. Nikon NIS-Elements D 3.10 Advanced image processing software was used to capture and analyze the images. The fluorescence signal from apoptotic cells was measured and the apoptotic rate was calculated as reported by us 58 .
Caspase activity assay. After treatment, caspase activities were measured using caspase-Glo reagents (Promega, USA) through cleavage of colorless substrates specific for caspase 3/7 caspase 8 and caspase 9 using a multifunction microplate reader (Berthold Technologies, Mithras LB940, Germany). In each assay, at least six wells per sample were measured for each dose and the results were averaged 59 .
Pgp assays. Functional analysis. Based on retention of rhodamine 123 in the whole zebrafish, we developed an in vivo assay for assessing drug effects on Pgp efflux in zebrafish (our published patent: China patent No. 201110231726.2). In these studies, we determined that rhodamine 123 staining for 40 min was the optimal time point for assessing Pgp efflux in untreated and drug treated animals. AB wild type zebrafish at 6 hpf were distributed into 6-well microplates (Nest, NEST Biotech), 30 zebrafish per well in 3 mL fish water. Furanodiene at 3 various concentrations (1/9 MTC, 1/3 MTC and MTC) was added to the treatment solution. After 24 h treatment, zebrafish were stained with rhodamine 123 for 40 min, then photographed under a fluorescence microscope. Nikon NIS-Elements D 3.10 software was used to quantify the whole zebrafish fluorescence intensity. (2019) 9:4541 | https://doi.org/10.1038/s41598-019-40866-2 www.nature.com/scientificreports www.nature.com/scientificreports/ Western blot. Total protein of compound-treated zebrafish was extracted with the Tissue Protein Rapid Extraction Kit (Keygen, China) and determined by performing the bicinchoninic acid (BCA) protein assay (Pierce, Waltham, MA). The equal amounts of 100 mg lysate proteins were loaded onto SDS-polyacrylamide gels (12%) at 80 V and electrophoretically transferred onto polyvinylidene difluoride membranes (Millipore, Billerica, MA) in Tris-glycine buffer at 100 V in an ice box. After blocking with 5% nonfat milk in TBST for 1 h at room temperature, the membrane was incubated with TNNT2 (1:1000, rabbit antibodies, Abcam, London, UK) overnight at 4 °C, After washed thrice with TBST, the membrane was incubated with a horseradish peroxidase-conjugated anti-rabbit Ig G secondary antibody (Abcam, Britain) for 1 h on a shaking table at room temperature, and washed thrice with TBST and the antibody-bound proteins were detected using the ECL chemiluminescence reagent (Pierce, Waltham, MA).
Statistical analysis. All data were presented as mean ± SE. Statistical analysis and graphical representation ofthe data were performed using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA). If a single concentration of a drug was assessed, the Student's t-test was used to identify drugs that exhibit a significant effect compared to the vehicle control group. If multiple concentrations of a drug were assessed, ANOVA was first used to assess whether there were any differences in the mean among the various concentrations of each compound; if a significant difference was determined (p < 0.05), Dunnett's test, which is appropriate for multiple pair-wise comparisons against a control, was then performed.

Data Availability
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.