Fucoxanthin and fucoxanthinol were obtained from WAKO Pure Chemical Co., Ltd. (Osaka, Japan). Fluorouracil and paclitaxel were obtained from Kyowa Hakko Kirin Co., Ltd. (Tokyo, Japan) and Nippon Kayaku Co., Ltd. (Tokyo, Japan), respectively. DF (10) medium (Dulbecco's modified Eagle' medium/F-12; Gibco, Grand Island, NY, USA) was supplemented with 10% FBS, 0.5 mg/ml piperacillin sodium (Toyama Chemical Co., Ltd., Tokyo, Japan), 0.25 mg/ml kanamycin sulfate (Meiji Seika Pharma Co., Ltd., Tokyo, Japan), 10 µg/ml vancomycin hydrochloride (Kobayashi Kako Co., Ltd., Fukui, Japan) and 1.25 µg/ml amphotericin B (Sumitomo Dainippon Pharma Co., Ltd., Osaka, Japan).

The DLD-1 (ATCC CCL-221), HCT116 (ATCC CCL-247), SW620 (ATCC CCL-227), Caco-2 (ATCC HTB-37), Colo205 (ATCC CCL-222) and WiDr (ATCC CCL-218) human colorectal cancer cell lines were obtained from the American Type Culture Collection (Manassas, VA, USA).

Surgically resected colorectal cancer specimens were obtained from 20 patients (12 males and 8 females; mean age, 68.5 years; age range, 39–82 years) from 2011 until 2013. Informed consent to measure fucoxanthin and fucoxanthinol sensitivities was obtained from all patients. The present study was approved by the ethics committee at Hakodate Municipal Hospital (Hokkaido, Japan).

CD-DST was performed as described previously by Kobayashi et al (11–13). Surgically resected tissue was cut into small pieces and washed several times with saline supplemented with 0.5 mg/ml piperacillin sodium, 0.25 mg/ml kanamycin sulfate, 10 µg/ml vancomycin hydrochloride and 1.25 µg/ml amphotericin B. The tissue sections were finely minced using a razor blade and digested in a cell dispersion enzyme solution (EZ; Kurabo, Osaka, Japan) for 2 h with rotation. The pellet was separated by centrifugation at 250 × g for 3 min. The supernatant was then removed and the pelleted cells were treated with 1 mM EGTA-0.03% trypsin for 3 min. DF medium was added by gentle pipetting and resuspended cells were filtered through a 300-µm nylon mesh. The collected cells were cultured in PCM-1 medium (Kurabo). Cells were then pre-incubated for 24 h at 37°C, and viable cells that adhered to the collagen gel were collected by dissolving the gel using collagenase (C0130; Sigma-Aldrich, St. Louis, MO, USA) in a flask containing collagen gel (CG-flask; Kurabo). Only viable cells were suspended in the reconstructed type I collagen solution (Cellmatrix Type CD; Kurabo) at a final density of 5×10⁴ − 2×10⁵ cells/ml. The collagen-suspended cells were dropped onto a 6-well plate, in order to prepare 3 drops of 30 µl each and were gelated at 37°C in a 5% CO₂ incubator for 1 h. DF medium was overlaid on each well, and the plate was incubated in 5% CO₂ at 37°C overnight. Fucoxanthin or fucoxanthinol (3 µl) solution dissolved in DMSO was added into the 3 ml of culture medium of each well, and the final concentrations were adjusted to 5, 10 or 20 µM. Plates were maintained at 37°C in the 5% CO₂ incubator with 100% humidity for 24 h. In addition, wells were prepared with fluorouracil or paclitaxel, at a final concentration of 1 µg/ml. Following incubation, DF medium containing the carotenoids or the conventional anticancer drugs was removed, and each well was rinsed twice with DF, which did not contain FBS, overlaid with serum-free culture medium PCM-2 (Kurabo), and incubated for 6 days. On the fourth day of incubation, the PCM-2 medium was replaced. At the end of the cultivation, 30 µl of Neutral Red (Kurabo) solution was added to each well at a final concentration of 50 µg/ml and viable cells were stained in a 5% CO₂ incubator at 37°C for 2 h. After removing PCM-2 medium containing Neutral Red, cells were fixed with 10% neutral formalin buffer, washed in water, air dried and quantified by image analysis (High-resolution Video Microscope VH-5910, Keyence, Osaka, Japan). The in vitro sensitivity to fucoxanthin, fucoxanthinol and the anticancer drugs is expressed as T/C (%), where T is the absorbance of cells which stained by neutral red treated with carotenoids and C is the absorbance of non-staining cells. A T/C (%) of ≤50% was considered to indicate efficacy of an agent in vitro. In addition, a T/C (%) reduction that was >2 standard deviations (T/C (%) <80%) was considered to be significant since the standard deviation of the control group of the clinical colorectal cancers was 100% ± 10% (data not shown).

The sensitivity of colorectal cancer cell lines to fucoxanthin and fucoxanthinol was estimated using CD-DST, which analyzes sensitivity to anticancer drugs and has been clinically applied in individualized chemotherapy (9–11). Treatment with 20 µM fucoxanthin decreased the T/C (%) of Caco-2, WiDr and HCT116 cell lines (Table I). Furthermore, fucoxanthinol decreased the T/C (%) of SW620 and DLD-1 cells in a similar manner to the three cell lines affected by fucoxanthin, although fucoxanthin did not affect the SW620 or DLD-1 cell lines. In all colorectal cancer cell lines, the T/C (%) of fucoxanthinol-treated cells was lower than that of fucoxanthin-treated cells at a dose of 20 µM. In particular, the T/C (%) values of Caco-2 and WiDr cells that were incubated for 6 days following 24 h treatment with 20 µM fucoxanthinol, markedly decreased to 1.4±0.2% and 12.0±0.3%, compared with 73.9±4.0% and 65.3±2.6%, respectively, for the same cell lines treated with fucoxanthin. By contrast, the T/C (%) of Colo205 cells did not decrease following treatment with either carotenoid. By contrast, the T/C (%) of Colo205 cells did not become less than 80% following treatment with fucoxanthin or fucoxanthinol.

Fucoxanthin and fucoxanthinol exhibited dose-dependent anticancer effects in colorectal cancer specimens (Table II). Fucoxanthinol markedly decreased the T/C (%) of colorectal cancer tissue samples. The proportion of samples treated with 20 µM fucoxanthinol, in which the T/C (%) values were <50%, was higher (13/20, 65%) than that in samples treated with fucoxanthin (2/20, 10%). Following treatment with 10 µM fucoxanthinol, the T/C (%) values of 5 samples were <50%, and 3 samples exhibited no change following treatment with fucoxanthin or fucoxanthinol. Further, the median T/C (%) value of the 20 cancer samples was 79.7% and 35.1% at 10 µM and 20 µM fucoxanthinol, respectively, while the median T/C(%) values for samples treated with 10 µM and 20 µM fucoxanthin were 94.6% and 80.6%, respectively.

The conventional anticancer drugs, fluorouracil and paclitaxel, also reduced the T/C (%) of colorectal cancer samples. Fluorouracil treatment resulted in T/C (%) values of samples from 52.3±6.8 − 113.4±5.5%. A T/C (%) value <50% was not observed in any of the samples treated with 1.0 µg/ml fluorouracil, which is the standard concentration administered to patients with colorectal cancer (11,12). Paclitaxel treatment resulted in T/C (%) values from 20.1±1.4 – 126.6±5.2%. T/C (%) values <50% were observed in 5 of 19 cases (26.3%) treated with 1.0 µg/ml paclitaxel. The median T/C (%) values of the 20 samples treated with fluorouracil and paclitaxel were 86.3 and 75.8%, respectively, which are approximately the same levels as those of samples treated with 20 µM fucoxanthin and 10 µM fucoxanthinol. Furthermore, treatment with 10 and 20 µM fucoxanthinol reduced the T/C (%) of drug-resistant colorectal cancer samples, including cases 3 and 4, in which fluorouracil and paclitaxel did not exert anticancer effects. In addition, case 2 had a diagnosis of malignant melanoma, based on the pathological findings from immunohistochemical staining that demonstrated distinct melanin granules. Fucoxanthin and fucoxanthinol also exhibited anticancer effects in the malignant melanoma sample, at doses of 20 and 10 µM, although fluorouracil and paclitaxel did not. Thus, the anticancer effects of fucoxanthin and fucoxanthinol on colorectal cancer samples differed from those of fluorouracil and paclitaxel.