This article is mentioned in:

Introduction

Pancreatic cancer is the 8th most common cause of cancer-related deaths worldwide due to its poor prognosis and difficulty of diagnosis and treatment (1). Mortality rates have remained constant over the past three decades. Chemotherapy is the mainstay treatment for the 15-20% of patients whose pancreatic tumours can be effectively removed surgically, as well as for patients with tumours that are not surgically resectable (2). However, >90% of pancreatic cancer cases are resistant to current chemotherapies (3,4). At present, the DNA synthesis inhibitor, gemcitabine, (2',2'-difluoro-2'-deoxycytidine) is used as the preferred chemotherapeutic agent (5,6), despite the fact that even when combined with other chemo- or radio-therapeutic agents, it exhibits limited efficacy, and produces severe adverse reactions (5,7). Gemcitabine can be toxic to healthy cells, resulting in unpredictable severe toxic effects. The inherent biological characteristics of pancreatic cancer, as well as the blood-pancreas barrier, possibly contribute to the unsatisfactory therapeutic effects of gemcitabine (8).

Thus, there is a need for more effective anticancer drugs which have lesser toxic side effects to improve clinical efficiency. In the present study, fucoxanthin, a natural product first isolated from the marine seaweeds Fucus, Dictyota, and Laminaria in 1914, and thus one of the most abundant carotenoids, was examined (9). Fucoxanthin is found in both macroalgae such as Undaria pinnatifida or Laminaria japonica, as well as microalgae such as Phaeodactylum tricornutum or Cylindrotheca closterium (10). Fucoxanthinol, the main fucoxanthin metabolite, has been detected in human plasma after the daily intake of wakame (Japanese common name for U. pinnatifida). Its bioavailability and metabolism are higher in humans than in mice (11) and does not exhibit any significant adverse effects in vivo (12). It has been regarded as a potential natural substance for cancer treatment (9). Although human experiments on the effects of fucoxanthin are lacking, animal experiments have shown evidence of anticancer effects. Oral administration of fucoxanthin has revealed no toxicity and mutagenicity (13-17). Apart from anticancer effects, fucoxanthin has also been demonstrated to have other health beneficial effects (18-22). Fucoxanthin uses various mechanisms to suppress tumour formation; inducing autophagy (21,22), arresting the cell cycle at the G1/G0 phase, inducing apoptosis, enhancing gap junctional intercellular communication, and involving different regulatory events in SAPK/JNK, Akt/mTOR, Bcl-2, JAK/STAT, MAPK and NF-κB pathways (23,24). Fucoxanthin has been shown to enhance chemotherapeutic efficacy of cisplatin in vitro (25). Recently, the combination therapy (chemotherapy in combination with fucoxanthin) has been evaluated in colon and liver cancers (25-29), lung cancer (30), breast cancer (31), and cervical cancer (32).

Pancreatic ductal adenocarcinoma accounts for >90% of all pancreatic cancers (33), and MIA PaCa-2 and PANC-1 cell lines, derived from primary tumours, are currently used as in vitro models to study pancreatic ductal adenocarcinoma carcinogenesis (34,35).

The aim of the present study was to investigate the suppressive effect of fucoxanthin on the growth of pancreatic cancer cell lines, MIA PaCa-2 and PANC-1, and the effect of combining it with gemcitabine. The 293 cell line, as a reference line for potential toxicity towards a non-cancer cell line, and MIA PaCa-2 and PANC-1 cell lines were utilised in the present study to determine the various cytotoxic effects of treatment between pancreatic cancer cells and a control. Flow cytometry was used to determine the possible mechanism of action by analysing the cell cycle and apoptosis.

Materials and methods

Materials

Fucoxanthin (cat. no. F6932) and MTT (cat. no. M2128) were purchased from Sigma-Aldrich NZ; Merck KgaA. Human pancreatic cancer cell lines, MIA PaCa-2 (cat. no. CRL-1420TM) and PANC-1 (cat. no. CRL-1469TM), and human cell line, 293 (cat. no. CRL-1573TM) were purchased from American Type Culture Collection (ATCC). Cell culture medium (RPMI-1640), penicillin-streptomycin and L-glutamine were purchased from Thermo Fisher Scientific, Inc. Fetal bovine serum was purchased from Moregate BioTech. All three cell lines were stored in liquid nitrogen. Following thawing, the cell lines were maintained in a tissue culture flask containing complete growth culture medium (RPMI-1640) with 1% of L-glutamine, 1% of penicillin-streptomycin and 10% of fetal bovine serum. All cells were cultured in an incubator at 37˚C, with 5% carbon dioxide humidified air.

MTT assay

MTT assays were used to determine cell viability (36). Cells were seeded at densities of 5,000 cells/well in 96-well plates, for 6-24 h and then cultured in the incubator at 37˚C for the following experiment. A total of 100 µl of fresh complete culture medium containing various concentrations of drugs (i.e., gemcitabine 25, 50 and 500 nM) was added to corresponding wells. Following incubation for 0, 24, 48 and 72 h, the medium was carefully removed and replaced with 100 µl of fresh complete culture medium. An aliquot of 10 µl of MTT stock solution (0.2 mg/ml) was added to each well and the plates were placed in the 37˚C incubator for 2 h. The supernatant was gently removed from the wells. An aliquot of 150 µl of DMSO was added to each well and mixed thoroughly using an orbit plate shaker. After incubating at 37˚C for 20-30 min, the plate was shaken briefly and the absorbance was measured by a plate reader (FLUOstar Omega; Alphatech Systems, Ltd.), at 540 nm. Background subtraction was utilised for generation of final values.

Determination of the effect of fucoxanthin

The cell density selected in this study for MIA PaCa-2, PANC-1 and 293 cells was 5,000 cells/well, according to our preliminary culture study. Following proper dilution, a multi-channel pipette was used to seed 100 µl of cells/well into the 96-well plate. The fucoxanthin stock solution was prepared by dissolving fucoxanthin in absolute ethanol with its concentration as 5 mM. Fucoxanthin (prepared in absolute ethanol) was diluted with cell culture medium. The range of concentrations of fucoxanthin were different for the three cell lines: MIA PaCa-2 (fucoxanthin concentration from 0.02 to 1 µM); PANC-1 (from 0.5 to 100 µM) and 293 (from 1 to 100 µM). Since fucoxanthin was dissolved in ethanol, the effect of ethanol alone on the cells was determined. The results revealed that 2, 1.6 and 1.0% fucoxanthin inhibited the growth of MIA PaCa-2 cells after incubating for 24, 48 and 72 h. Ethanol at a concentration of ≤0.5% did not significantly affect the growth of MIA PaCa-2 cells. No effect on the growth of PANC-1 cells cultured with ethanol controls was noted. The study was then carried out using fucoxanthin concentrations under 25 µM, and 100 µl of each diluted fucoxanthin solution was added to wells immediately for analysis.

Determination of the colour effect of fucoxanthin on the absorbance value (OD value)

As fucoxanthin is an orange-coloured pigment, its colour may impact the final absorbance value of each well when performing MTT assays. Different concentrations of fucoxanthin solution were prepared to assess respective absorbance values. The highest fucoxanthin concentration used was 100 µM and two test methods were used for this study. The first was performed by testing 100 µl of each fucoxanthin solution under the plate reader directly; and the other method was carried out by following the MTT assay protocol. A total of 10 µl MTT solution was added to each well and incubated at 37˚C for 4 h, then 150 µl DMSO was added and finally the absorbance was read at 540 nm. According to the results, fucoxanthin solution was replaced with fresh cell culture medium before the addition of the MTT solution in the present study.

Determination of optimum concentration of gemcitabine

According to previous single treatment experiment results, the optimum concentrations of fucoxanthin and gemcitabine were determined. In the combination treatment experiment, fucoxanthin at concentrations of 150, 250 and 300 nM, and gemcitabine, at 25 and 50 nM were concurrently combined with each other to treat MIA PaCa-2 cells. In addition, fucoxanthin concentrations (10 and 20 µM) were combined with gemcitabine (50 and 500 nM, respectively) to treat PANC-1 cells. An MTT assay was applied to determine the cell viability, and the experimental steps were the same as those aforementioned.

Synergism analysis

The nature of the combination between gemcitabine and fucoxanthin was quantified by synergism quotient (SQ) (37-39). SQ is defined as the net growth inhibitory effect of the combination treatment by the sum of the net individual treatment effect on growth inhibition. A quotient of >1.1 indicates a synergistic effect, a quotient between 0.9 and 1.1 indicates an additive effect, while a quotient of <0.9 indicates an antagonistic effect (38).

Cell cycle assay

After detaching, counting and diluting, cells were seeded onto 6-well plates. The seeding density for all cell lines in this study was 50,000 cells/ml and a 2-ml cell solution was seeded in each well. The plates were maintained in a 37˚C incubator for 6-24 h to ensure that almost all of the cells were detached from the walls of the wells.

When the cell attachment rate was high, cells were treated with 2 ml of serum-free medium (no FBS, with 1% penicillin and 1% L-glutamine) in each well, for synchronizing cell proliferation. One plate set in this study was a control group and the other plates were designed as treatment groups. In Plate 1 (control group), wells A, B and C were treated with 2 ml of serum-free medium (no FBS), and wells D, E and F were filled with 2 ml of complete culture (10%) medium. Before adding the fresh medium (serum-free medium or 10% FBS medium), the old medium was removed first. The fresh medium was added slowly and gently in circles along the wall of the well. For Plate 2 (treatment group), all the wells were treated with 2 ml of serum-free medium. The plates were maintained in the 37˚C incubator for 24 h.

All treatments were maintained in 15 ml centrifuge tubes which were prepared with complete culture (10% FBS) medium. The old serum-free medium was first removed gently, and then 2 ml of well-mixed treatment solution was carefully added into the designated wells. A total of 2 ml complete culture medium was added to the control well. All plates were maintained in the incubator for 48 h (day 2) and 72 h (day 3). Control cells were collected from Plate 1 with day 0 samples. Namely, fucoxanthin, 150, 250 and 300 nM, and gemcitabine, 25 and 50 nM, were used to treat MIA PaCa-2 solely and jointly. For PANC-1 cells, fucoxanthin concentrations were 10 and 20 µM and gemcitabine were 50 and 500 nM.

Cell cycle analysis

The permeabilizing solution of (total volume) 1 ml per test tube consisted of: 0.1% Triton X-100 (1 µl for each 1 ml) + RNAse A (50 µg/ml from stock solution of 1 mg/ml). For example, 20 tubes were formulated and dispensed as follows: 20 µl Triton X-100 + 1,000 µl RNAse +19,980 µl PBS. After centrifuging all tubes at 125 x g for 2 min at 0-4˚C, the ethanol was gently removed, and 3 ml ice cold PBS was added to each tube before the second centrifugation step. Following the second centrifugation (125 x g for 2 min at 0-4˚C), another 3 ml of ice-cold PBS was added to each tube to replace the initial wash. In total, the cells were washed twice with 3 ml ice cold PBS. In the process of PI staining, the supernatant in each well was gently removed and then 1 ml of well-mixed permeabilizing solution was added to each tube. All cells in the tubes were carefully mixed before being transferred to test tubes and then incubated at 37˚C for 30-45 min. Following permeabilization, 5 µg/ml of PI (5 µl to 1 ml in each tube) was added to each test tube and incubated for 5 min. Finally, all the tubes were analysed under a flow cytometer (MoFlo XDP Beckman Coulter, Inc.).

The IC50 values were calculated using PRISM® software (version 6.0; GraphPad Software, Inc.), and the IC50 values were obtained using dose-response inhibition, nonlinear regression (curve fit): Log (inhibitor) vs. response-variable slope (four parameters). Kaluza® Flow Cytometry Analysis Software (version 1.3; Beckman Coulter, Inc.) was applied in cell cycle results analysis to measure the cell cycle distributions in the present study.

Statistical analysis

All experiments in the present study were performed at least three times. Data were assessed for normal distribution before analysis. Statistical differences in multiple groups were determined by one-way analysis of variance (ANOVA) with PRISM® software (version 6.0; GraphPad Software, Inc.) and SPSS (version 22.0; IBM Corp.). Statistical comparisons were performed using Tukey's post hoc test. Analysis between two groups was determined using unpaired Student's t-test. Data are expressed as the means ± standard deviation (SD; sample n=3 with triplicate analysis performed on each sample). P<0.05 was considered to indicate statistically significant differences.

Results

In order to analyse the treatment effect of gemcitabine or fucoxanthin on pancreatic cancer cell lines MIA PaCa-2, PANC-1, and 293 cell growth, cells were treated in vitro in a dose or time-and-dose course experiment. Experiments were performed in triplicate and repeated at least three times independently (Table I).

Table I Cytotoxicity (IC50) of treatment in MIA PaCa-2 cells, PANC-1 and 293 cells detected at various time points: gemcitabine (72 h) or fucoxanthin (24, 48 and, 72 h).

Table I

Cytotoxicity (IC50) of treatment in MIA PaCa-2 cells, PANC-1 and 293 cells detected at various time points: gemcitabine (72 h) or fucoxanthin (24, 48 and, 72 h).

	IC50 (µM) aftergem-citabine treatment	IC50 (µM) after fucoxanthin treatment
Cell line	72 h	24 h	48 h	72 h
MIA PaCa-2	25.00±0.47	17.72±1.04	10.68±0.64	8.74±0.28
PANC-1	48.55±2.30	N/A	N/A	10.58±0.56
293 Cells	48.82±3.27	28.97±1.58	8.70±1.17	8.28±0.30

[i] Data are presented as the mean ± SD (n=6). IC₅₀, half maximal inhibitory concentration.

Inhibitory effect of gemcitabine or fucoxanthin on MIA PaCa-2 pancreatic cancer cells

Culture of cells with various concentrations of gemcitabine for 72 h resulted in significant suppression of cell viability in a dose-dependent manner. ANOVA analysis revealed that cell viability was statistically and significantly decreased with the increasing concentration of gemcitabine, thus the concentration of gemcitabine was a significant factor in altering cell viability (P<0.01); the IC50 was 25.00±0.47 nM after 72 h of treatment (Table I). Gemcitabine at 25 and 50 nM exhibited approximately 63.45 and 42.18% cell viability of MIA PaCa-2 cells at 48 h, respectively (Table II). These two concentrations were selected in the following combination study.

Table II Cell viability percentages of MIA PaCa-2 cells after 72 h of incubation with gemcitabine, fucoxanthin, or gemcitabine-fucoxanthin combination.

Table II

Cell viability percentages of MIA PaCa-2 cells after 72 h of incubation with gemcitabine, fucoxanthin, or gemcitabine-fucoxanthin combination.

	Gemcitabine (nM)
Compound Fucoxanthin (nM)	0 (%)	25 (%)	50 (%)
0	100	63.45±0.54	42.18±0.48
150	99.8±0.19	59.76±1.63a	39.81±0.69a
250	98.99±0.89	53.06±0.70a	38.32±0.52a
300	99.83±0.61	50.40±0.50a	36.72±0.12a

[i] Data are presented as the mean ± SD (n=6).

[ii] ^aP<0.01 (Student's t-test).

Fucoxanthin inhibited the proliferation of MIA PaCa-2 cells in a concentration-dependent manner (Table II) at 72 h. One-way ANOVA revealed that cell viability was statistically and significantly decreased with the increasing concentration of fucoxanthin treatment (P<0.01) after treatment for 72 h. The IC50 values within the three days of treatment significantly and statistically decreased to 17.72±1.04 (24 h), 10.68±0.64 (48 h) and 8.74±0.28 µM (72 h) (Table I). Morphological changes observed microscopically became more marked with increasing culture time, in the presence of fucoxanthin concentrations >6.25 µM. Cells diminished in size and were scattered and more easily detached. After 72 h, the volume of cells was significantly reduced, and the edges of the cells were rough (data not shown).

Inhibitory effect of gemcitabine or fucoxanthin on PANC-1 pancreatic cancer cells

Gemcitabine was effective in inhibiting PANC-1 cells at only 72 h of incubation in a dose-dependent manner, yielding an IC50 value of 48.55±2.30 nM (Table I). ANOVA revealed that cell viability was statistically and significantly decreased with the increasing concentration of gemcitabine treatment, thus, the alteration of treatment concentration was associated with a change in cell viability.

Furthermore, fucoxanthin inhibited proliferation of PANC-1 cells in a dose-dependent manner after 72 h incubation. Only 72-h incubations were performed in view of the aforementioned result of gemcitabine. The IC50 value was 10.58±0.56 µM (Table I). ANOVA revealed that treatment with fucoxanthin was a significant factor in the inhibition of cell growth with increasing treatment concentrations and exposure times.

Inhibitory effect of gemcitabine and fucoxanthin on 293 cell line

The same gemcitabine concentrations used for MIA PaCa-2 cells were used for 293 cells. Gemcitabine significantly inhibited cell viability in a concentration-dependent manner. The IC50 value was 48.82±3.27 nM (72 h) (Table I). Fucoxanthin inhibited 293 cell growth at concentrations in a dose- and time-dependent manner. The IC50 values showed a significant decrease after three days of treatment to 28.97±1.58 (24 h), 8.70±1.17 (48 h), and 8.28±0.30 µM (72 h), respectively (Table I). ANOVA indicated that the treatment with gemcitabine or fucoxanthin was a significant factor in the inhibition of cell growth with increasing treatment concentrations and exposure times.

Inhibitory effect of gemcitabine-fucoxanthin combination

Based on single drug experiments, 72 h of incubation time for combination drug treatment was used for MIA PaCa-2 cells. The cell viability in groups treated with combination of gemcitabine and fucoxanthin for 72 h, is presented in Table II. The cell viability was not markedly altered with fucoxanthin treatment alone (concentrations from 0 nM to 300 nM). However, gemcitabine 25 nM combined with 150, 250 and 300 nM fucoxanthin significantly reduced the cell viability by approximately 4, 10 and 13%, respectively, as compared with gemcitabine alone (cell viability, 63.45±0.54%). Similarly, in the gemcitabine 50-nM treatment group, the enhanced inhibitory effect of the combination treatment with fucoxanthin was also observed: Gemcitabine treatment 50 nM alone (42.18±0.48%) vs. gemcitabine treatment with increasing concentrations of fucoxanthin at 150 (39.81±0.69%), 250 (38.32±0.52%) and 300 nM (36.72±0.12%). According to the statistical analysis, single treatment with fucoxanthin alone could not significantly decrease the cell viability (P>0.01) at treatment concentrations of 150, 250 and 300 nM. However, gemcitabine treatment alone at 25 and 50 nM significantly decreased the cell viability (P<0.01). Combination treatment also significantly decreased the cell viability (P<0.01). Additionally, in the presence of fucoxanthin, the cell viability was significantly decreased in a dose-dependent manner as compared to gemcitabine treatment alone (P<0.01). As revealed in Table III, treatment with gemcitabine at 25 nM with increasing concentrations of fucoxanthin from 150 to 300 nM, yielded an SQ value which increased from 1.079 (additivity) to 1.316 (synergism). Gemcitabine, at 50 nM, with increasing concentrations of fucoxanthin, increased the SQ value from 1.034 (additivity) to 1.085 (additivity). These results indicated that treatment with fucoxanthin generated a synergistic effect at a combined treatment with 25 nM of gemcitabine. In all groups, with increasing concentrations of fucoxanthin, the cell viability of MIA PaCa-2 cells decreased. Hence, fucoxanthin was able to enhance the inhibitory effect of gemcitabine on cell viability of MIA PaCa-2 cells in a concentration-dependent manner, even at a low concentration range. The single treatment results revealed that PANC-1 cells were not sensitive to gemcitabine as compared to MIA PaCa-2 cells. Therefore, gemcitabine at 50 and 500 nM was used in combination with fucoxanthin at 10 and 20 µM. The treatment time was also 72 h.

Table III Synergism analysis of MIA PaCa-2 cells after 72 h of incubation with combination treatment of gemcitabine-fucoxanthin.

Table III

Synergism analysis of MIA PaCa-2 cells after 72 h of incubation with combination treatment of gemcitabine-fucoxanthin.

	Gemcitabine (nM)
Compound Fucoxanthin (nM)	0	25	50
0	N/A	N/A	N/A
150	N/A	1.079	1.034
250	N/A	1.205	1.033
300	N/A	1.316	1.085

[i] Data are presented as the mean ± SD (n=6).

The PANC-1 cell viability values under different concentrations of drugs are listed in Table IV. After 72 h of incubation, with 10 and 20 µM fucoxanthin alone, the number of viable cells decreased to 56.91±3.00 and 18.54±1.10%, respectively. Treatment with gemcitabine alone decreased the number of viable cells. However, even at a concentration of gemcitabine increased 10 times from 50 to 500 nM, the cell viability was minimally affected (from ~10 to 18%, respectively). This effect was observed in all combination treatments. In the combination treatment group (fucoxanthin, 10 µM), the addition of gemcitabine at concentrations of 50 and 500 nM did not significantly reduce (~3%) the viability of cancer cells. Once increased, a fucoxanthin concentration of 20 µM, demonstrated only a 2% increase in the inhibition rate at gemcitabine concentrations of 50 and 500 nM. The single treatment of either fucoxanthin alone or gemcitabine alone decreased the cell viability. However, the combination treatment did not generate a significant effect for reduced cell viability (P>0.01). As noted in Table V, the concentration of fucoxanthin (from 10 to 20 µM) increased the SQ values in this stated range: 50 nM gemcitabine [0.889 (antagonistic effect) to 0.902 (additive effect), respectively] and 500 nM [0.823 (antagonistic effect) to 0.840 (antagonist effect)].

Table IV Cell viability percentages of PANC-1 cells after 72 h of incubation with gemcitabine, fucoxanthin, or gemcitabine-fucoxanthin combination.

Table IV

Cell viability percentages of PANC-1 cells after 72 h of incubation with gemcitabine, fucoxanthin, or gemcitabine-fucoxanthin combination.

	Gemcitabine (nM)
Compound Fucoxanthin (nM)	0 (%)	50 (%)	500 (%)
0	100	90.20±0.67	82.15±0.88
10	56.91±3.00a	52.92±0.63a	49.42±2.51a
20	18.54±1.10b	17.42±1.20b	16.01±0.36b

[i] Data are presented as the mean ± SD (n=6).

[ii] ^aP<0.01 and

[iii] ^bP<0.001 (Student's t-test).

Table V Synergism analysis of PANC-1 cells after 72 h of incubation with combination treatment of gemcitabine-fucoxanthin.

Table V

Synergism analysis of PANC-1 cells after 72 h of incubation with combination treatment of gemcitabine-fucoxanthin.

	Gemcitabine (nM)
Compound Fucoxanthin (nM)	0	50	500
0	N/A	N/A	N/A
10	N/A	0.889	0.823
20	N/A	0.902	0.840

[i] Data are presented as the mean ± SD (n=6).

Gemcitabine significantly inhibited the cell viability of 293 cells by approximately 23, 49 and 76% after 72 h of incubation at 25, 50 and 500 nM concentrations, respectively (Table VI; P<0.05). In the fucoxanthin alone groups (150, 250, 300, 10,000 and 20,000 nM), there was no significant suppression by fucoxanthin at ≤300 nM, but there was significant inhibition at high concentrations of 10 and 20 µM, as compared to the control (P<0.01). In the combination treatment groups, adding fucoxanthin at concentrations of 0 to 150 nM, slightly increased cell viability (2.58 and 4.05%; P<0.05) at gemcitabine treatment concentrations of 25 and 50 nM, respectively. No significant differences in cell viability were observed at gemcitabine treatment concentrations of 25 and 50 nM combined with fucoxanthin treatment concentrations of 150 to 300 nM (P<0.01). However, combination with high concentrations of fucoxanthin (10 and 20 µM), induced significantly greater inhibition than the gemcitabine alone group (P<0.01). This may be due to the cytotoxicity of fucoxanthin used alone at high concentrations (Table VI). In Table VII, gemcitabine (25 nM) had an SQ value increase from 0.833 (antagonistic effect) to 1.087 (additive effect) as the concentration of fucoxanthin increased from 150 to 300 nM. Consistently, the SQ value increased from 0.900 (additive effect) to 0.918 (additive effect) with increasing fucoxanthin concentrations of 150 to 300 nM in the gemcitabine 50-nM groups. However, once the concentration of fucoxanthin increased to 10 mM, the SQ value decreased below 0.700 (antagonistic effect).

Table VI Cell viability percentages of 293 cells after 72 h of incubation with gemcitabine, fucoxanthin, or gemcitabine-fucoxanthin combination.

Table VI

Cell viability percentages of 293 cells after 72 h of incubation with gemcitabine, fucoxanthin, or gemcitabine-fucoxanthin combination.

		Gemcitabine (nM)
Compound Fucoxanthin (nM)	0 (%)	25 (%)	50 (%)	500 (%)
0	100	77.42±2.34a	51.27±0.88b	23.60±1.79c
150	99.08±1.71	80.00±5.80	55.32±2.20
250	99.06±1.23	79.02±4.49	55.80±2.49
300	100.04±1.99	75.92±4.77	55.73±2.31
10000	30.51±2.92b		22.47±2.30b	20.04±1.45
20000	12.72±0.66c		11.25±1.03c	9.89±0.93b

[i] Data are presented as the mean ± SD (n=6).

[ii] ^aP<0.05,

[iii] ^bP<0.01 and

[iv] ^cP<0.001 (Student's t-test).

Table VII Synergism analysis of 293 cells after 72 h of incubation with combination treatment of gemcitabine-fucoxanthin.

Table VII

Synergism analysis of 293 cells after 72 h of incubation with combination treatment of gemcitabine-fucoxanthin.

	Gemcitabine (nM)
Compound Fucoxanthin (nM)	0	25	50	500
0	N/A	N/A	N/A	N/A
150	N/A	0.833	0.900
250	N/A	0.875	0.900
300	N/A	1.087	0.918
10000	N/A		0.655	0.559
20000	N/A		0.649	0.565

[i] Data are presented as the mean ± SD (n=6).

Combination of gemcitabine and fucoxanthin on the cell cycle. The cell cycle distribution of MIA PaCa-2 cells is presented in Fig. S1. According to Fig. S1 the G0-G1 phase was significantly increased from 41.04% (control) to 60.37% (24-h serum-starved cells) and the G2-M and S phases were decreased from 27.00% (control) to 15.75% (24-h serum-starved cells) and 17.28% (control) to 10.62% (24-h serum-starved cells), respectively. The distribution of the cell cycle was not altered by incubation with 150 or 250 nM fucoxanthin alone for either 48 or 72 h (Table SI). However, fucoxanthin at a concentration of 300 nM has a slight effect: The percentage of the G0-G1 phase increased ~1.9 and 3.75% at both 48 and 72 h, respectively. Treatment with 10 µM fucoxanthin for 48 h, resulted in an increase of MIA PaCa-2 cells in G0-G1 phase (from 48.78 to 59.51%), and the percentage of sub-G1 was increased (from 8.0 to 12.83%). At 72 h, fucoxanthin (10 µM) not only increased the cells in sub-G1 phase, but also increased the percentage of cells in the G2-M phase with a concomitant decrease in the number of cells in the S phase (Table SI). Gemcitabine (at 25 and 50 nM) caused a significant accumulation of cells in the S phase, as compared to the control at both 48 and 72 h, while the number of cells in the G0-G1 phase was reduced (Table SI). Additionally, sub-G1 phase accumulation increased in a time- and dose-dependent manner, particularly with 50-nM gemcitabine treatment at 72 h.

In the combination treatment groups, the results indicated that fucoxanthin (150, 250 and 300 nM) combined with gemcitabine in MIA PaCa-2 cells, for 48 h, increased the percentage of cells in the S phase, but the cells in the sub-G1 phase were not altered (Table SI), as compared to gemcitabine (25 and 50 nM) alone (Fig. 1A). At the 72-h combination treatment, the sub-G1 percentage was enhanced with the corresponding increase in fucoxanthin concentration. However, sub-G1 phase accumulation in the 25-nM gemcitabine combination groups was more apparent than the 50-nM gemcitabine groups (Fig. 1B).

Figure 1 Effects of fucoxanthin on gemcitabine-induced cell cycle arrest in MIA PaCa-2 cells. (A) Percentage of MIA PaCa-2 cells in the S phase after treatment for 48 h. (B) Percentage of MIA PaCa-2 cells in the sub-G1 phase after treatment for 72 h. Data are presented as the mean ± SD (n=3). **P<0.01 (ANOVA with post hoc Tukey's test).

Effects of gemcitabine and fucoxanthin on PANC-1 cell cycle progression

The cell cycle distribution of PANC-1 cells is presented Table SII. In addition, cell starvation (medium-only, without serum) for 24 h, synchronised and blocked the PANC-1 cells in G0-G1 phase (Fig. S2). The percentage of cells in the G0-G1 phase was increased by ~19%, while the number of cells in the S phase decreased from 20.73 to 9.62% and the number of cells in the G2-M decreased from 24.34 to 18.03%. In the groups treated only with fucoxanthin (10 or 20 µM) for 48 h, the results revealed that the accumulation of cells in the G0-G1 phase was significantly increased compared to the control (Table SII). Furthermore, the percentage of cells in the S phase decreased with the increase of fucoxanthin concentration. At 72 h, fucoxanthin 20 µM did not block the cells in the G0-G1 phase. However, fucoxanthin induced the increase of the percentage of cells in the sub-G1 phase (24.70%) compared with the control cells (9.00%) at 72 h (Table SII). Fucoxanthin decreased the proportion of cells in the G2-M phase and increased the percentage of sub-G1 cells in a time- and dose-dependent manner. Gemcitabine, 50 nM, 500 nM and 50 µM blocked the cells in the G0-G1 phase at 48 h, and the percentage of cells in the sub-G1 phase was increased in a time- and dose-dependent manner. Similar results were obtained at 72 h, except for gemcitabine at a concentration of 50 µM which induced cell apoptosis [56.65 (control) vs. 25.99% (50 µM gemcitabine); Table SII)].

As revealed in Table SII, in the combination treatment groups (50 and 500 nM gemcitabine combined with 10 µM fucoxanthin), an increase in the percentage of cells in the G0-G1 phase as well as in the sub-G1 phase was observed, compared with the cells treated only with 10 µM fucoxanthin for 48 h. The accumulation of cells in the S phase increased with higher gemcitabine concentrations (50 to 500 nM), and changes in the G2-M phase were negligible. However, cells treated with the combination drugs for 72 h did not exhibit increased arrest of the G0-G1 phase, with respect to cells treated only with fucoxanthin 10 µM. By contrast, the percentage of cells in the sub-G1 phase increased after treatment for 48 and 72 h (Fig. 2A and B). As for cells treated with 20 µM fucoxanthin combined with gemcitabine (50 and 500 nM), the combined treatment induced increased accumulation of cells in the S phase and cell apoptosis, as compared to cells treated only with 20 µM fucoxanthin, at both at 48 and 72 h (Table SII). G2-M phase detection was negligible in the fucoxanthin 20-µM combination treatment groups at both time points, similar to the 10-µM fucoxanthin combination treatment groups (Table SII).

Figure 2 Effects of fucoxanthin on gemcitabine-induced cell cycle arrest in PANC-1 cells. (A) Percentage of PANC-1 cells in the sub-G1 phase after treatment for 48 h. (B) Percentage of PANC-1 cells in the sub-G1 phase after treatment for 72 h. Data are presented as the mean ± SD (n=3). **P<0.01 (ANOVA with post hoc Tukey's test).

Effects of gemcitabine and fucoxanthin on 293 cell cycle progression

The results of cell cycle distribution revealed that 24-h starvation (medium-only, without serum) did not arrest 293 cells in the G0-G1 phase; it was demonstrated that the percentage of serum-starved cells in the G0-G1 phase was slightly decreased compared with the control (Fig. S3). The cell cycle distribution of 293 cells is presented in Table SIII. In the single treatment groups, the distribution of cell cycle phases in fucoxanthin-treated (150, 250 and 300 nM) 293 cells was similar to that in control cells at both 48 and 72 h after treatment. Cells treated only with fucoxanthin at a concentration of 10 µM at 48 h induced a 1.53% higher cell apoptosis value than the control cells, and the percentage of cells in the G0-G1 phase was even lower than that of control cells. The G2-M and S phases were not markedly altered in relation to the controls. However, at 72 h, 10 µM fucoxanthin was responsible for a higher accumulation of cells in the G2-M phase, and its percentage of sub-G1 cells was only ~2% higher than the same concentration at 48 h. There were no significant differences in the cell cycle distribution of 25-nM gemcitabine-treated cells, at 48 h, in relation to the control, and apoptosis in sub-G1 was increased by only ~6%. However, treatment with 50-nM gemcitabine induced apoptosis from sub-G1 (25.28%), and most of the cells were arrested at the S phase. At 72 h, gemcitabine at a concentration of 25 nM increased the percentage of sub-G1 over a small range as compared with control group. Gemcitabine, at a concentration of 50 nM, induced cell apoptosis (sub-G1) of up to 38.57% and no cell phase distribution could be observed, implying a high level of apoptotic induction in these cells (Table SIII).

In the combination treatment groups [fucoxanthin (150, 250 and 300 nM) combined with gemcitabine (25 and 50 nM), for 48 and 72 h], no significant differences in the distribution of all phases in the cell cycle were revealed, as compared to cells treated only with gemcitabine. Combination treatments induced a slightly lower percentage of cell apoptosis compared to that observed with single-gemcitabine treatment.

Discussion

To the best of our knowledge, this is the first study, investigating the effects of fucoxanthin and its combination with gemcitabine in pancreatic cancer cells. The characteristics of high anticancer efficacy, with mild or no toxicity to normal cells, is the primary objective in cancer research. Fucoxanthin used alone in treatment has exhibited significant anticancer efficacy (31), however, when used in combination with gemcitabine as determined in the present study, was more effective with less harmful effects to normal cells.

The results obtained in the present study indicate that the inhibitory effect of gemcitabine on MIA PaCa-2, PANC-1 and 293 cells occurs in a dose-dependent manner. This is consistent with previous studies of gemcitabine on, pancreatic ductal adenocarcinoma cell lines MIA PaCa-2 and PANC-1 (39,40), breast adenocarcinoma cell lines MCF-7 and MDA-MB-231(41), colorectal adenocarcinoma cell line HT-29(42), and cervical carcinoma cell lines (43). The IC50 values for gemcitabine determined in this study were 25.00±0.47 nM for MIA PaCa-2 cells, 48.55±2.30 µM for PANC-1 cells and 48.82±3.27 nM for 293 cells. These three different IC50 values indicated a different sensitivity of the three cell lines to gemcitabine. MIA PaCa-2 cell line was the most sensitive to treatment of gemcitabine. PANC-1 and 293 cell lines both showed strong chemo-resistance with 293 cells displaying slightly stronger chemo-resistance than PANC-1. Previous studies also revealed the same results as concluded in this study (38,39,44). PANC-1 was found to have low sensitivity to gemcitabine in a previous study (40). MIA PaCa-2 and PANC-1 are both primary tumour cells (35). The difference in the IC50 values between these cell lines could be due to the fact that PANC-1 cells are more resistant as compared to MIA PaCa-2 cells. Moreover, in a previous study it was reported that even at very low doses (5-10 nM), gemcitabine was able to induce nuclear factor-κB (NF-κB) activity, which can promote gemcitabine chemo-resistance (45). Hence, it can be surmised that doses (>50 nM) used in this study could significantly activate the NF-κB activity.

In the present study, fucoxanthin inhibited the viability of human pancreatic cancer cell lines, MIA PaCa-2 and PANC-1, as well as 293 cells in a dose-dependent manner. Fucoxanthin time-dependently suppressed the proliferation of MIA PaCa-2 and 293 cells. In the higher fucoxanthin concentration treatment groups of each cell line, formation of nuclear condensation was evidently clear when observed under the inverted microscope (data not shown). The antioxidant property of fucoxanthin was considered to be one of the major reasons for the anticancer effect of fucoxanthin (46-48). NF-κB activity has been demonstrated to be inhibited by antioxidants (49,50). It has been suggested that some part of fucoxanthin is hydrolysed to fucoxanthinol during the uptake (51). Fucoxanthinol is then further converted into amarouciaxanthin A in HepG2 cells. These two fucoxanthin metabolites were found to reduce the viability of human prostate cancer cell line PC-3 (27,52). Fucoxanthinol was also found to have more efficient anti-proliferative effects than fucoxanthin (23,53,54). The difference in sensitivity for different cell lines may be due to the different content of hydrolytic enzymes in the different cells.

In the present investigation, gemcitabine was used with fucoxanthin simultaneously, to explore their combined effects on pancreatic cancer cells. To the best of our knowledge, applying a fucoxanthin concentration range under 1 µM in anticancer research was performed for the first time. The aim of using low fucoxanthin concentrations was to determine whether fucoxanthin is able to effectively improve the cytotoxicity of gemcitabine even at low concentrations. The carotenoid fucoxanthin is known to sensitize multidrug-resistant cancer cells, and a proposed mechanism for fucoxanthin overcoming multiple drug resistance in cancer cells and increasing efficacy of chemotherapy in targeted cells has been reported (55).

The findings of the present study may potentially increase the benefits of fucoxanthin-coupled treatments. The combined effect of fucoxanthin and gemcitabine on the reduction of MIA PaCa-2 cell viability was significantly higher than either gemcitabine (25 and 50 nM) or fucoxanthin (150, 250 and 300 nM) used alone. A synergistic effect was only observed in the 25-nM gemcitabine combined with fucoxanthin at concentrations higher than 250-nM groups. In addition, gemcitabine (50 nM) with fucoxanthin only demonstrated an additive effect.

A possible explanation for this observation may be that gemcitabine at a concentration of 50 nM is highly cytotoxic, and therefore low doses of fucoxanthin cannot provide additional effects beyond the chemotherapeutic dose. PANC-1 cells were found to be resistant to gemcitabine and sensitive to fucoxanthin treatment. Thus, fucoxanthin (10 and 20 µM) played a leading role in the combination effect in the suppression of PANC-1 cells. Fucoxanthin was shown to significantly enhance the inhibitory effect of gemcitabine. The cell viability under the combination treatment was significantly lower than that in cells treated only with fucoxanthin or gemcitabine. However, the combination effect was not synergistically improved by these two drugs.

In the present study, it was found that low doses of fucoxanthin treatment (150, 250, 300 nM) could not inhibit the growth of 293 cells for 72 h. However, high concentrations of the fucoxanthin (10 and 20 µM) significantly suppressed the proliferation of 293 cells. Low concentrations of fucoxanthin did not promote the inhibitory effect of gemcitabine (25 nM). Interestingly, in the 50-nM gemcitabine groups, low doses of fucoxanthin aided in reducing the cytotoxicity of gemcitabine. In addition, there was more 293 cell survival under the incubation with combination treatment for 72 h in relation to cells cultured only with gemcitabine (50 nM). Fucoxanthin was demonstrated to have no inhibitory effect on human lymphocyte cells, uninfected leukaemia cell lines and human peripheral blood mononuclear cells, over a certain concentration range (23). These literature studies potentially indicate that fucoxanthin is still toxic to non-cancer cells when used at high doses, although its cytotoxicity is selective. Through its mechanism of action, gemcitabine has been determined as a pyrimidine nucleotide analog to be involved in DNA synthesis, thereby inhibiting the DNA synthesis and relating to cell division from the whole process of cell mitosis (56-59). Conversely, fucoxanthin exerts it anti-proliferative and cancer-preventing effects via different molecules and pathways including the Bcl-2 proteins, MAPK, apoptosis, or metastasis (51). Kumar et al reported that the anti-proliferative effects of fucoxanthin are selective, i.e., fucoxanthin has the capability to target cancer cells only, leaving normal physiological cells unaffected or less affected (51). Therefore, similar to the results of this previous study, the results of the present study indicate that fucoxanthin selectively exerts its effect only on pancreatic cancer cells.

A suitable concentration range of fucoxanthin and gemcitabine has to be established in order to create synergistic effects in cancer cell growth inhibition and neutralize toxicity to non-cancer cells. The arrest of cell cycle progression, induction of apoptosis, or both, are factors in the inhibition of cancer cell proliferation (60). The anticancer activity of gemcitabine is primarily performed by impairing DNA synthesis. It results in the cytostasis owing to the block of the cell cycle in the G0-G1 or S phases (61). Subsequently, cells may undergo apoptosis or mitotic catastrophe upon escaping the cell cycle blockage, which will finally lead to cell death (62). Gemcitabine, depending on its exposure time and concentration, has been demonstrated to block certain human solid tumour cells in the S phase (63). However, a previous study indicated that gemcitabine arrests cells in the G0-G1, G1, early S or S phases only depending upon the concentration of gemcitabine (64). In the present study, gemcitabine was demonstrated to dose-dependently arrest the MIA PaCa-2 cells in the S phase after both 48- and 72-h exposure. Sub-G1 is an index of apoptotic DNA fragmentation (65). The results of the present study indicated that gemcitabine first blocked MIA PaCa-2 cells in the S phase and then induced cell apoptosis. The same result was obtained in 293 cells. However, gemcitabine was revealed to arrest PANC-1 cells in the G0-G1 phase after culture for 48 and 72 h. Gemcitabine induced the apoptosis of PANC-1 cells in a time- and dose-dependent manner. Taken together, the results of the present study on MIA PaCa-2, PANC-1 and 293 cells are consistent with previous research aforementioned, namely that gemcitabine induces G0-G1 and S phase arrest and subsequently undergoes apoptosis. Moreover, all these results are consistent with the results of the cytotoxicity analysis.

Carotenoids have been demonstrated to inhibit tumour cell growth by inducing cell cycle arrest at the G1 phase and/or apoptosis (66,67). Fucoxanthin has been suggested to accumulate cells in the G0-G1 phase of different cell lines as previously reported (68). In the present study, higher concentrations of fucoxanthin were found to arrest the cells in the G0-G1 phase. Fucoxanthin has been observed to block the human gastric adenocarcinoma cell line, MGC-803, in the G2-M phase (69). By contrast, it has been suggested that carotenoids cannot arrest cells in the G2-M phase (70). Thus, whether fucoxanthin induces the arrest of cells in the G2-M phase requires further research. Low doses of fucoxanthin (150, 250 and 300 nM) could not markedly alter the cell cycle distribution. This is consistent with the cytotoxicity analysis.

Fucoxanthin combined with gemcitabine was found to help induce pancreatic cancer cell arrest in the G0-G1 or S phase. The lack of cell cycle arrest by fucoxanthin on 293 cells implies that fucoxanthin has selective toxicity. Both gemcitabine and fucoxanthin block cells in the G1/S phase and the effects of each compound on cellular metabolism are different. From literature reported, carotenoids reverse multidrug resistance and enhance sensitivity of cancer treatment in in vitro and in vivo models (55,71,72). However, in the present study only additive effects were observed. A plausible explanation for this observation may be the sequence of drug administration. The administration of scheduled medications has been demonstrated to be very important for the antitumor effect. The same drugs with different treatment sequences were found to produce differing results (73). Gemcitabine and fucoxanthin were added simultaneously to the cell lines in the present study. Therefore, pre-treating the cells with one of the compounds and later treating these cells with the second one should be assessed in further studies. Nevertheless, low doses of fucoxanthin were revealed to help improve the anti-proliferative efficacy of gemcitabine by inducing growth arrest, while high doses of fucoxanthin inhibited proliferation by inducing apoptosis. The synergistic effect could be observed by using the best combination concentrations and optimal treatment sequence of fucoxanthin and gemcitabine.

In conclusion, fucoxanthin effectively improved the cytotoxicity of gemcitabine even at low concentrations, and as a type of carotenoid, fucoxanthin generated a synergistic effect increasing the sensitivity of chemotherapy in certain pancreatic cancer cell lines. Fucoxanthin also exhibited selective toxicity against cancer cells even at low concentrations. Thus, fucoxanthin, the most abundant carotenoid found in marine algae, at low therapeutic concentrations, may be considered as a potential adjunct treatment for pancreatic cancer in combination with other clinical cancer chemotherapy drugs. A limitation of the present study is identified in the MTT assay. This assay was used as a single method for cell viability determination, while other methods may also be applied. The color of fucoxanthin itself may be a potential risk that could affect the actual result of the colorimetric assay. Hence other cell viability assays such as Cell Counting Kit-8 or crystal violet assays may be used to validate the cytotoxic results. In addition, using non-cancerous pancreas-origin cell lines may be a more appropriate choice for future study. Therefore, the present results are not conclusive. Furthermore, more data should be generated, such as changes in cell morphology, and mechanism studies detailing the effect of fucoxanthin on caspase-3 or caspase-9 should be performed to demonstrate the effectiveness of fucoxanthin treatment. In addition, the selectivity of observed fucoxanthin bioactivity should be examined using in vivo studies.

Supplementary Material

Cell cycle distribution of MIA PaCa-2 treated with gemcitabine in the presence and absence of fucoxanthin.

Cell cycle distribution of PANC-1 cells after treatment with gemcitabine in the presence and absence of fucoxanthin.

Cell cycle distribution of 293 cells after treatment with gemcitabine in the presence and absence of fucoxanthin.

Cell cycle distribution of MIA PaCa-2 cells after treatment with gemcitabine in the presence and absence of fucoxanthin.

Cell cycle distribution of PANC-1 cells after treatment with gemcitabine in the presence and absence of fucoxanthin.

Cell cycle distribution of 293 cells after treatment with gemcitabine in the presence and absence of fucoxanthin.

Acknowledgements

Not applicable.

Funding

Funding: The present study was supported by the New Zealand-China Tripartite Partnership Fund (to JLu, JLi, BZ and TY) of the New Zealand Ministry of Education, the Royal Society of New Zealand Catalyst Seeding Fund (grant no. 21-AUT-005-CSG), and the Shanghai Engineering Research Center of Plant Germplasm Resources (grant no. 17DZ2252700).

Availability of data and materials

The data generated in the present study may be requested from the corresponding authors.

Authors' contributions

JLu, JLi, TY, BZ and TF conceived the study. JLu, KSW, XJW, YL, LC, YZ, MJ, JLiu designed the study and acquired the data. AH, YH, JLu, XJW and KSW analyzed the data. JLu, JLi, TY, BZ and TF provided experimental materials. XJW, AH, KSW and JLu wrote the original draft. YL, LC, YZ, MJ, JLiu, JLu, JLi, TY, BZ, TF and YH revised the work critically for important intellectual content. TF, KSW and JLu confirm the authenticity of all the raw data. All authors read and approved the final manuscript and agree to be accountable for all aspects of the research in ensuring that the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References