Anticancer effects of the combined Thai noni juice ethanolic extracts and 5-fluorouracil against cholangiocarcinoma cells in vitro and in vivo

Application of 5-fluorouracil (5-FU) in cholangiocarcinoma (CCA) is limited by adverse side effects and chemoresistance. Therefore, the combination therapy of 5-FU with other substances, especially natural products may provide a new strategy for CCA treatment. The aim of this study was to evaluate the combination effects of 5-FU and two ethanolic extracts of Thai noni juice (TNJ) products on CCA cell lines and nude mice xenografts. The results of antiproliferative assay showed the combination treatment of 5-FU and each TNJ ethanolic extract exerted more cytotoxicity on CCA cells than either single agent treatment. Synergistic effects of drug combinations can enable the dose reduction of 5-FU. The mechanism underlying a combination treatment was apoptosis induction through an activation of p53 and Bax proteins. In the nude mouse xenograft model, combination treatments of 5-FU with each TNJ ethanolic extract suppressed the growth of CCA cells implanted mice more than single agent treatments with no effects on mouse body weight, kidney, and spleen. Moreover, low doses of TNJ ethanolic extracts reduced the hepatotoxicity of 5-FU in nude mice. Taken together, these data suggested that the ethanolic extracts of TNJ products can enhance the anti-CCA effect and reduce toxicity of 5-FU.

of its low cost. Thus, the development of less toxic and more effective anticancer agents in combination with 5-FU is urgently needed. Accordingly, natural products are an interesting source of several phytochemicals with potential to be chemotherapeutic agents. Nowadays, the combination therapy of chemo-natural products is considered as promising.
Morinda citrifolia L., commonly known as noni and called Yor in Thailand, is a traditional plant in tropical Asia that is widely used as a functional food and medicinal herb. Noni has been reported to have many pharmacological properties, such as antimicrobial, antifungal, antioxidant, anti-arthritic, anti-inflammatory and anticancer activity 10 . Currently, noni juice has been used as an herbal drink and is commercially available worldwide. Noni fruit juice is a natural source of numerous bioactive compounds, including phenolic acids, flavonoids, iridoids and coumarin 11 . Previous research revealed that the noni juice does not induce adverse effects in hepatocytes, and renal function or show subchronic oral toxicity 12,13 . Moreover, several reports suggest that noni juice has anticancer properties and combination treatment with noni juice can improve response rate and reduce the toxicity of current chemotherapy drugs [14][15][16] . However, the effects of noni juice in combination treatment with chemotherapeutic agents on CCA have not been reported.
To date, Thai noni juice (TNJ) products are distributed in the local market and worldwide. Several brands of TNJ product are produced from different regions which have been approved by the Food and Drug Administration of Thailand. Although the commercial TNJ has been claimed to have several health benefits, detailed studies on the anticancer effects in combination treatment with chemotherapy drugs have not been reported. In our previous study, two commercial ethanolic extracts of TNJ products, Thai noni juice Chiangrai (TNJ-Cr) and Thai noni juice Buasri (TNJ-Bs), induced apoptosis in two CCA cell lines, KKU-100 and KKU-213B, by up-regulating the expression of Bax through a p53-dependent mechanism 17 . The partial phenolic profiles of both TNJ ethanolic extracts were evaluated by HPLC and the identified phenolic acids included gallic, protocatechuic, p-hydroxybenzoic, vanillic, syringic, p-coumaric, and ferulic acids. Some major phenolic acids including p-hydroxybenzoic, vanillic, and protocatechuic acids were identified in both TNJ ethanolic extracts, however, the highest HPLC peaks of both extracts remain to be identified. In this study, the anti-CCA effects of single and combination agent treatments between 5-FU and two TNJ ethanolic extracts (TNJ-Cr or TNJ-Bs) were evaluated on CCA cell lines and on nude mice xenografts. The mechanisms underlying their synergism were also determined. This finding may provide a new strategy for CCA treatment.
Antiproliferative effects of the combination treatment of 5-FU and TNJ ethanolic extracts. The combined effects of 5-FU with the various concentrations of each TNJ ethanolic extract on CCA and non-cancer cells were determined. The results of synergistic effects between combined treatments of 5-FU and TNJ-Cr or TNJ-Bs ethanolic extracts on KKU-100, KKU-213B and H69 cells are shown in Fig. 2. Combination treatments of 5-FU and TNJ ethanolic extracts at 48 h significantly inhibited CCA cell proliferation more than either single agent treatment at the synergistic concentrations for 48 h. As shown in Fig. 2a, cell viability of KKU-100 significantly decreased to 48.55 ± 3.29% or 47.26 ± 5.81% in the combined treatments of 5-FU and TNJ-Cr or TNJ-Bs ethanolic extracts for 48 h, respectively. While 5-FU alone reduced KKU-100 viability to 79.66 ± 3.05% at 48 h of treatment. The ethanolic extracts of TNJ-Cr or TNJ-Bs alone reduced KKU-100 viability at 48 h incubation to 94.04 ± 1.68% or 89.78 ± 3.94%, respectively. Likewise, 5-FU combined with TNJ-Cr or TNJ-Bs ethanolic extracts significantly reduced cell viability of KKU-213B at 48 h to 52.91 ± 8.11% or 45.43 ± 8.56%, respectively, that was more than either single agent treatment at the synergistic concentrations for 48 h (64.18 ± 1.78% for 5-FU, 88.10 ± 6.32% for TNJ-Cr ethanolic extract and 80.39 ± 6.70% for TNJ-Bs ethanolic extract) (Fig. 2b). To elucidate the toxicity of the combined treatments on non-cancer H69 cell, the highest concentration of the combined agents of 5-FU with TNJ-Cr or TNJ-Bs ethanolic extracts that resulted in synergism in CCA cells were used to treat H69. In Fig. 2c

Effect of the combination treatment of 5-FU and TNJ ethanolic extracts on cell cycle distribution and apoptosis induction.
To investigate the cell cycle arrest and apoptosis induction in the combination treatment of 5-FU and TNJ ethanolic extracts, the drug resistant KKU-100 cells were chosen as they showed the strongest synergism for the combination treatments. The dose and exposure time of combined drugs that caused the synergism with 50% inhibition of cell proliferation was used to conduct the cell cycle analysis and apoptosis assay in KKU-100 cell. As shown in Fig. 3a,b, the treatment of 5-FU alone at IC 20 (75 µM) increased the populations in sub-G1 23 ± 3.39% and in S phase 12.65 ± 0.49%, while the ethanolic extracts of TNJ-Cr or TNJ-Bs alone at the synergistic concentration did not affect any cell cycle phase in KKU-100, but slightly increased the sub-G1 population (6.15 ± 0.63% for 0.25 mg/mL TNJ-Cr and 8.05 ± 0.49% for 1.00 mg/mL TNJ-Bs). In the combination treatment, 5-FU combined with TNJ ethanolic extracts did not cause cell cycle arrest but instead   . The data are represented as mean ± SD of three independent experiments. *p < 0.05 indicates a significant difference between the treatments and the solvent control; **p < 0.05 indicates a significant difference between the single and combined agent treatments.   Table S1). No mice died in any group after CCA cell inoculation and treatment with the single or combined agents (Fig. 5b). Mice were sacrificed and the tumors were excised and photographed (Fig. 5c). As shown in Fig. 5d-f, mice groups treated with TNJ-Cr or TNJ-Bs ethanolic extracts alone in the dose of 125 or 250 mg/kg had a decrease in tumor volume and weights after 2 weeks of treatment when compared to the groups treated with the vehicle control and with the 5-FU alone. Moreover, the groups treated with combined 5-FU and TNJ ethanolic extracts showed a greater decrease in tumor volume. The results showed no significant difference between groups treated with the low or high doses of TNJ ethanolic extracts within either single agent or combined regimes. However, the mean tumor volume and weight in groups treated with combined 5-FU and a high dose TNJ ethanolic extracts were less than for tumors from groups treated with combined 5-FU and a low dose TNJ ethanolic extracts.   www.nature.com/scientificreports/ assay were used to further examine tumor histology and apoptosis induction in the mice (Fig. 6). Histopathology analysis of tumors in the groups treated with the single and combined agents of 5-FU and TNJ ethanolic extracts showed nuclear condensation in the tumor cells (Fig. 6a). Compared to the tumors in the vehicle control group, the TUNEL-positive cells with brown-stained nuclei were increased in groups treated with 5-FU or TNJ ethanolic extracts alone (Fig. 6b). Moreover, the tumors of groups treated with combined 5-FU and TNJ ethanolic extracts showed more intense staining of TUNEL-positive cells than the tumors of groups treated with either agent alone. Furthermore, in the groups treated with combined 5-FU and a high dose TNJ ethanolic extract (250 mg/kg) showed significantly increased TUNEL stained cells compared to groups treated with combined 5-FU and low dose TNJ ethanolic extracts (125 mg/kg) (Fig. 6c).
In vivo toxicological evaluation. As shown in Table 2 and Supplementary Table S3, mean initial and final body weights of mice showed no significant changes between the vehicle control and the treatment groups.
In addition, liver weights were not affected by the single agent treatment with the low or high doses of TNJ ethanolic extracts when compared to the vehicle control group. The mice with 5-FU alone or in combination with a high dose TNJ ethanolic extracts (250 mg/kg) had significantly reduced in liver weights. In contrast, the combined treatment of 5-FU and a low dose of TNJ ethanolic extracts (125 mg/kg) did not significantly change liver weights. Compared to the vehicle control group, the treatment groups with 5-FU, TNJ-Cr and TNJ-Bs ethanolic extracts alone or in combination showed no significant effects on kidney and spleen weights. Histological sections of mouse livers revealed normal hepatocyte morphology in the groups treated with the vehicle control or the single agent of TNJ-Cr or TNJ-Bs ethanolic extracts with low or high dose (Fig. 7a). Whereas the groups treated with 5-FU alone or with the combination of 5-FU and a high dose TNJ-Cr or TNJ-Bs ethanolic extracts demonstrated hepatocyte injury as shown by the vacuolar or hydropic degeneration in the hepatocytes. However, the combination treatment of 5-FU with low dose TNJ ethanolic extracts showed no hepatocellular damage (Fig. 7a). Moreover, there were no significant differences in histological findings of mouse kidney and spleen tissues between the single or combined agents treated groups and the vehicle control (Fig. 7b,c).

Discussion
Chemotherapy of CCA with 5-FU is often associated with adverse side effects and drug-resistance. So, the development of 5-FU combination therapy with another substance such as a natural bioactive compound from a plant or herb extract is an alternative treatment that will enhance the anticancer activity and reduce the drug resistance of 5-FU [18][19][20]  www.nature.com/scientificreports/ extracts alone was considered as a chemotherapeutic resistance response of the cancer cells 28 . However, 5-FU in combination with TNJ ethanolic extracts could reduce the Bcl-2 level in KKU-100 cells when compared to the treatment with 5-FU alone. Moreover, the increasing Bax/Bcl-2 ratio in the combination treatment of 5-FU and TNJ ethanolic extracts indicated increasing apoptosis in the KKU-100 cells 29 . Our findings suggest that the TNJ ethanolic extracts decrease drug resistance of 5-FU treatment with the implication that Bcl-2 is a promising target to overcome the drug resistance in CCA cells. The experimental basis for the preclinical application of 5-FU in combination with TNJ ethanolic extracts was examined in nude mouse CCA xenograft model. Our results demonstrate that TNJ-Cr and TNJ-Bs ethanolic extracts enhance tumor suppression of 5-FU in the nude mice. Consistent with the in vitro results, apoptosis is the main effect of the combination treatment of 5-FU and TNJ ethanolic extracts on KKU-100-inoculated mice. Although no significant difference was observed in the mouse tumor volume between the 5-FU combined treatments of low (125 mg/kg) and high doses (250 mg/kg) of TNJ ethanolic extracts, the apoptotic-positive cells in the mouse tumors were increased in the combined group treated with 5-FU and high dose TNJ ethanolic extracts. This result suggests that a longer treatment time of more than 14 days may show a significant change of tumor volume between the combined treatments of 5-FU with low and high doses of TNJ ethanolic extracts. However, the average inhibition ratio of tumors in the group treated with 5-FU and a high dose TNJ ethanolic extract was greater than the group treated with 5-FU and a low dose TNJ ethanolic extract. In the in vivo toxicological evaluations, our results demonstrated that the mice treated with 5-FU and TNJ ethanolic extracts alone or in combination showed no significant adverse effects on body weights. However, the mean body weight of the mice treated with 5-FU alone was the lowest among the treatment groups. The mice liver weights were decreased in treatment groups with 5-FU alone or in the combination with a high dose of TNJ ethanolic extracts. Long term administration of 5-FU can lead to liver damage 30 and the changing liver weight reflects the loss of liver mass due to the metabolism of 5-FU in the liver 31,32 . So, the liver damage in nude mice may be related to the side effects of 5-FU administration. However, both TNJ ethanolic extracts alone did not affect the mice liver weight. The mice treated with 5-FU and a low dose of TNJ-Cr or TNJ-Bs ethanolic extracts showed significant recovery of liver weights as well as normal hepatocytes when compared to mice treated with 5-FU alone. Therefore, hepatotoxicity in mice may be due to the 5-FU administration, and a low dose of TNJ ethanolic extracts appears to reduce the toxicity of 5-FU. Herbal medicine can reduce liver damage of 5-FU through eliminating free radicals and reduced oxidative damage from 5-FU toxicity 33 . Moreover, our findings demonstrate that TNJ ethanolic extracts alone and in combination treatment with 5-FU did not affect mouse kidney and spleen. Taken together, the TNJ ethanolic extracts enhanced the anticancer activity of 5-FU in the treatment of the drug resistant CCA cells. In addition, a low dose of TNJ ethanolic extracts appeared to reduce the toxicities of chemotherapy. However, clinical studies on the anticancer activity and the toxicity of the combination treatment of 5-FU and TNJ ethanolic extracts in the CCA patients still require further investigation.
In our previous study, the phenolic acid contents of both TNJ ethanolic extracts were partially identified based on the availability of phenolic acid standards and the fact that phenolic acids in several plant extracts possessed anticancer activity in our previous findings 34,35 . Methanolic extracts of peanut testae composed of predominantly p-hydroxybenzoic and p-coumaric acids had the antiproliferative activity against cervical, colon, leukemia and breast cancer cells 34 . Moreover, p-coumaric, ferulic, and sinapinic acids possessed histone deacetylase inhibitory activity against breast and cervical cancer cell lines 35 . Partial identification of phenolic acid composition in TNJ ethanolic extracts revealed that p-hydroxybenzoic, vanillic, and protocatechuic acids were the major phenolic acids in ethanolic extracts of TNJ 17 , which were reported to be natural anticancer agents [36][37][38] .
Yoshitomi et al. identified seven major compounds, asperulosidic acid, deacetylasperulosidic acid, scopoletin, morindolin, and three fatty acid glycosides, in the noni juice and they demonstrated that deacetylasperulosidic acid had an activity to reduce the blood pressure 39 . However, the anticancer activity of deacetylasperulosidic

Groups Initial body weight (g) Final body weight (g)
Organ weight (g/100 g body weight) www.nature.com/scientificreports/ acid has not yet been reported. Some other identified compounds had been reported the anticancer activity [40][41][42][43][44] . Asperulosidic acid possessed the antitumorigenic effects through inhibition of AP-1 transactivation and cell transformation in the mouse epidermal JB6 cell line 40 . Scopoletin exerted the anticancer effects by triggering apoptosis, cell cycle arrest and inhibition of cell invasion 41 . Rutin reduced cell viability in the liver cancer cells 42 . Moreover, other minor active compounds in noni extracts were also reported to possess the anticancer activity such as quercetin 43 , kaempferol 43 , and chrysin 44 . The anticancer activity of TNJ ethanolic extracts is dependent on the amount and type of active compounds in the extracts, however, these compounds may be active as single or in cooperative contribution with their natural cocktails presented in the extracts. For example, dichloromethane extract of fresh noni leave showed more antiproliferative activity than the pure compounds rutin or scopoletin, suggesting that an individual compound from the plant has lost the synergistic effect of various natural ingredients 45 . Nonetheless, the existence of other promising active compounds in the two TNJ ethanolic extracts will be further investigated before the clinical trial. www.nature.com/scientificreports/ In conclusion, combination treatments of 5-FU and TNJ-Cr or TNJ-Bs ethanolic extracts showed the synergistic antitumor effects on the human CCA cell lines, KKU-100 and KKU-213B. These synergistic effects can enable the dose reduction of 5-FU 5.84-and 12.86-fold for 50% cell proliferation inhibition in KKU-100 and KKU-213B cells, respectively. Apoptosis is the main mechanism for the synergism of 5-FU with TNJ ethanolic extracts on a drug resistant KKU-100 cells via the up-regulation of pro-apoptotic protein p53 and Bax. Moreover, the combined treatment of 5-FU and TNJ-Bs ethanolic extract was not toxic to the non-cancer H69 cells in vitro. Additionally, TNJ ethanolic extracts enhanced the antitumor growth of 5-FU on the mouse KKU-100 xenograft model through apoptosis induction. The treatment of 5-FU alone or in combination with high dose TNJ ethanolic extracts in KKU-100 inoculated nude mouse exerted toxic effects on mouse liver; whereas, combined treatment of 5-FU with low dose TNJ ethanolic extracts reduced the toxicity of 5-FU. Moreover, the combination treatment of 5-FU with TNJ ethanolic extracts shows no toxicity on mouse kidney and spleen. Therefore, the TNJ ethanolic extracts could be used in combination with 5-FU to reduce the toxicity and enhance the anticancer activity of the main drug. In a further study, the combination treatment of 5-FU with TNJ ethanolic extracts will be investigated in clinical trials.

Preparation of TNJ ethanolic extracts. Two commercial TNJ products (TNJ-Cr and TNJ-Bs) utilized
in this study were purchased from the market and the same lot number was used throughout the study. Thai noni juice Chiangrai (TNJ-Cr) and Buasri (TNJ-Bs) were obtained from Chiangrainoni Co., Ltd., Thailand (https:// openc orpor ates. com/ compa nies/ th/ 0575553000683), and Maebuasri brand of One Tumbon One Product (OTOP), Thailand (http:// www. nonib uasri. com), respectively. The same lot numbers of both TNJ products were used throughout this study. TNJ products (30 mL) were stirred continuously with 70 mL of absolute ethanol at room temperature for 4 h. The extracts were filtered through No.1 Whatman filter paper and then ethanol was Determination of drug interaction. The drug interactions of 5-FU with either TNJ-Cr or TNJ-Bs ethanolic extracts were determined by calculating the CI according to the Chou-Talalay method 49 . For 50% growth inhibition, the equation of CI values is as follow: CI = (D 1 /Dx 1 ) + (D 2 /Dx 2 ) + α⋅(D 1 ⋅D 2 )/(Dx 1 ⋅Dx 2 ), where D 1 is a dose of drug 1 (5-FU) in combination with drug 2 (TNJ ethanolic extracts) to produce 50% cell viability; Dx 1 is a dose of drug 1 alone to produce 50% cell viability; D 2 is a dose of drug 2 in combination with drug 1 to produce 50% cell viability; Dx 2 is a dose of drug 2 alone to produce 50% cell viability; α = 1 for mutually non-exclusive modes of drug action. CI < 1 shows a synergistic effect; CI = 1 shows an additive effect, and CI > 1 shows antagonism. The fold of dose reduction in combination treatment was indicated by DRI that provided the level of effect as compared to the dose of a single agent. The DRI was calculated using the following equation: DRI = Dx/D.

Cell cycle analysis.
To identify the mechanism under synergistic anti-CCA effects, cell cycle phase distribution was analyzed by using propidium iodide (PI) staining (Sigma Aldrich, St. Louis, MO, USA) and flow cytometry. Cells were seeded in a 5.5 cm dish plate at a density of 1 × 10 6 cells and cultured for 24 h. Then, the cells were treated with 5-FU and TNJ ethanolic extracts alone or in combination at the synergistic concentration for 48  and immunoreactive bands were exposed to X-ray film. The relative intensity was analyzed by the ImageJ program and using total ERK1/2 as a loading control in western blot analysis.  Histopathology. The tumor, liver, kidney and spleen were fixed in 10% formalin, dehydrated, cleared in tissue processing and embedded in paraffin. The samples were cut to 4 µm thickness by microtome and placed on a glass slide. The tissue sections were deparaffinized in xylene, rehydrated in ethanol (99, 95 and 70%), and then washed with distilled water. The rehydrated tissue sections were stained with hematoxylin and eosin. Subsequently, the samples were analyzed under a light microscope.
In situ apoptosis detection. Apoptotic cells of tumor tissue were detected by using an in situ cell death www.nature.com/scientificreports/ Statistical analysis. Data are expressed as means ± standard deviation (SD) from three independent experiments. Statistical analysis was performed using the statistical program IBM SPSS version 19.0 for windows (SPSS Corporation, Chicago, IL, USA) and Graphpad Prism version 8. One-way analysis of variance (ANOVA) and Dunnett's post-hoc test were used to analyze the statistical significance for multiple groups and Student's t-test was used to analyze the statistical significance between two groups. The statistical significance between different groups was considered to be at p < 0.05.

Data availability
The datasets generated and/or analyzed during the study are available from the corresponding author on reasonable request.