2020 年 43 巻 1 号 p. 49-52
Few therapeutic options exist for gemcitabine-resistant pancreatic cancer. In this study, we investigated the anti-cancer effects of 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors and bisphosphonates in pancreatic cancer cell lines (SUIT-2 and MIA PaCa-2) which show poor responses to gemcitabine, established through long-term culture in nutrient-deprived or gemcitabine-containing media. Under the nutrient-deprived condition, IC50s for statins and bisphosphonates decreased and those for gemcitabine increased compared with those under normal conditions. In cells cultured long-term with gemcitabine, although IC50s for gemcitabine increased, those for statins and bisphosphonates either slightly increased or remained unchanged. Thus, these drugs may be effective against pancreatic cancer cells which show poor responses to gemcitabine.
Pancreatic cancer has a very poor prognosis. Although gemcitabine-based chemotherapy is widely used for the treatment of pancreatic cancer, few therapeutic options exist in case the tumor develops gemcitabine resistance. Cancer cells may develop resistance to chemotherapy under nutrient-deprived conditions such as in the interior of large tumors. Previous studies on cultured cell lines have reported that pancreatic carcinomas cultured under nutrient-deprived conditions, such as low glucose or low amino acids, show resistance to chemotherapeutic agents, including gemcitabine.1,2) Furthermore, it has been reported that long-term exposure to gemcitabine confers gemcitabine resistance to pancreatic cancer cell lines.3,4) Our laboratory previously reported that fluvastatin, which is a 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitor, and/or zoledronate, which is a bisphosphonate, both induce anti-proliferative effects in cultured pancreatic cancer cells.5) In this study, we investigated whether statins and bisphosphonates possess anti-cancer effects against pancreatic cancer cell lines which show poor responses to gemcitabine, established either through long-term exposure to gemcitabine or by culturing under nutrient-deprived conditions.
Human pancreatic cancer cell lines MIA PaCa-2 and SUIT-2 were a kind gift from Dr. Soichi Takiguchi of National Kyushu Cancer Center (Fukuoka, Japan). For experiments on growth in a nutrient-deprived conditions, cells were grown in either a nutrition-rich medium (NRM) or a nutrition-deprived medium (NDM) in a humidified atmosphere of 5% CO2 at a constant temperature of 37°C for a few weeks. The NRM comprised RPMI-1640 media (Sigma-Aldrich Co. LLC., MO, U.S.A.) (containing amino acids and glucose) supplemented with 10% fetal bovine serum (FBS), penicillin, and streptomycin. The NDM comprised 60% reductions of amino acids, glucose, and FBS from the composition of the NRM. For experiments on growth in high-gemcitabine (Combi-Blocks Inc., CA, U.S.A.) conditions, cell lines show poor responses to gemcitabine were generated through long-term culture in gemcitabine-containing NRM. The concentration of gemcitabine was gradually increased by 10 nM per week from 10 nM to achieve a final concentration of 100 nM for MIA PaCa-2 cells and 200 nM for SUIT-2 cells. Standard NRM without gemcitabine was selected for healthy control cell lines (normal cell lines).
Cell Viabilities and IC50sThe cells were seeded into 96-well plates (Thermo Fisher Scientific Inc., MA, U.S.A.) at a density of 2.0 × 103 cells/well and used for experiments on the following day. The cells were exposed to each drug for 72 h, and cell viabilities were assessed through the mitochondrial activity represented by the reduction of WST-8 [2-(2-meth-oxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt] to formazan. To perform these tests, WST-8 assay solutions (Cell Counting Kit-8; Dojindo Laboratory, Kumamoto, Japan) were added to the cell cultures, and the cultures were further incubated for 1 h at 37°C in humidified air supplemented with 5% CO2. The extent of formazan dye formation was determined by measuring the absorbance at 450 nm using a microplate reader (iMark™ Microplate Reader; Bio-Rad Laboratories, Inc., CA, U.S.A.). IC50 values were calculated using approximated sigmoidal curves.6)
Statistical AnalysesData are shown as mean ± standard error of the mean (S.E.M.). Statistical analysis was performed using Student’s t-test (Statview; Abacus Concepts, CA, U.S.A.) to determine differences between the groups. A probability level of p < 0.05 was considered as statistically significant.
Chemotherapeutic agents including gemcitabine; oxaliplatin and paclitaxel; statins (simvastatin, atorvastatin, rosuvastatin, fluvastatin, and pitavastatin), except pravastatin; and bisphosphonates (zoledronate and alendronate) all demonstrated dose-dependent decrease in the cell viabilities of SUIT-2 cells cultured in both the NRM and NDM (Figs. 1A–H, J, K). However, pravastatin did not affect cell viability under either condition (Fig. 1I). Cells cultured in NDM containing gemcitabine (1 nM) and paclitaxel (1 and 10 nM) exhibited significantly higher cell viabilities than those cultured in NRM (Figs. 1A, C). Conversely, statins, except pravastatin, and bisphosphonates caused lower viabilities in cells cultured in NDM than in those cultured in NRM (Figs. 1D–H, J, K). IC50 values for almost all statins and bisphosphonates were smaller for cells cultured in NDM than for those cultured in NRM, although IC50 values for gemcitabine and paclitaxel were larger for cells cultured in NDM than for those cultured in NRM, in both SUIT-2 and MIA PaCa-2 cell lines (Table 1).
Cells were incubated with gemcitabine (A), oxaliplatin (B), paclitaxel (C), simvastatin (D), atorvastatin (E), rosuvastatin (F), fluvastatin (G), pitavastatin (H), pravastatin (I), zoledronate (J), and alendronate (K) for 72 h. Cell viabilities were measured using the WST-8 method. Open and closed circles mean the cell viabilities of cell lines cultured in NRM and NDM, respectively. Results are expressed as mean ± S.E.M. (n = 6–12). * p < 0.05, ** p < 0.01 compared with NRM.
| IC50 | ||||||||
|---|---|---|---|---|---|---|---|---|
| SUIT-2 | MIA PaCa-2 | SUIT-2 | MIA PaCa-2 | |||||
| Nutrient-rich medium (NRM) | Nutrient-deprived medium (NDM) | Nutrient-rich medium (NRM) | Nutrient-deprived medium (NDM) | Normal cell line | Cell line chronic-cultured with GEM | Normal cell line | Cell line chronic-cultured with GEM | |
| Chemotherapy agents | ||||||||
| Gemcitabine (nM) | 2.25 ± 0.29 | 4.58 ± 0.36** | 4.64 ± 0.39 | 8.90 ± 0.59** | 2.74 ± 0.48 | 12.8 ± 1.3** | 8.77 ± 0.47 | 10.2 ± 0.1* |
| Oxaliplatin (µM) | 9.04 ± 0.74 | 8.37 ± 0.66 | 20.8 ± 2.7 | 3.64 ± 0.96** | 8.44 ± 0.71 | 7.13 ± 0.87 | 20.8 ± 2.7 | 26.1 ± 4.4 |
| Paclitaxel (nM) | 7.92 ± 1.03 | 11.6 ± 1.1* | 0.518 ± 0.024 | 0.778 ± 0.085* | 5.47 ± 1.26 | 4.32 ± 0.93 | 0.518 ± 0.024 | 0.535 ± 0.021 |
| HMG-CoA reductase inhibitors | ||||||||
| Simvastatin (µM) | 7.88 ± 0.48 | 2.01 ± 0.49** | 0.947 ± 0.040 | 0.211 ± 0.092** | 10.2 ± 0.8 | 13.0 ± 0.9* | 0.947 ± 0.040 | 1.02 ± 0.05 |
| Atorvastatin (µM) | 11.4 ± 0.7 | 8.41 ± 1.12* | 1.31 ± 0.07 | 0.802 ± 0.154** | 15.7 ± 1.7 | 36.0 ± 3.6** | 1.31 ± 0.07 | 1.29 ± 0.11 |
| Rosuvastatin (µM) | 79.6 ± 5.2 | 67.2 ± 12.0 | 5.04 ± 0.37 | 3.77 ± 0.25* | 79.6 ± 5.2 | 120 ± 7** | 8.46 ± 1.01 | 6.60 ± 0.93 |
| Fluvastatin (µM) | 9.09 ± 0.46 | 2.41 ± 0.46** | 2.39 ± 0.40 | 0.508 ± 0.031* | 9.68 ± 0.47 | 13.2 ± 1.3** | 1.12 ± 0.04 | 1.05 ± 0.06 |
| Pitavastatin (µM) | 1.80 ± 0.13 | 0.963 ± 0.045** | 0.250 ± 0.009 | 0.245 ± 0.034 | 1.82 ± 0.08 | 4.14 ± 0.52** | 0.250 ± 0.009 | 0.265 ± 0.010 |
| Pravastatin (µM) | >100 | >100 | >100 | >100 | >100 | >100 | >100 | >100 |
| Bisphosphonates | ||||||||
| Zoledronate (µM) | 9.96 ± 1.25 | 2.28 ± 0.41** | 3.86 ± 0.96 | 1.14 ± 0.15* | 11.4 ± 1.3 | 18.9 ± 5.9 | 0.444 ± 0.033 | 0.363 ± 0.033 |
| Alendronate (µM) | 18.0 ± 2.1 | 5.94 ± 0.22 | 26.8 ± 1.9 | 1.83 ± 0.48** | 39.9 ± 7.4 | 38.7 ± 3.5 | 15.0 ± 1.4 | 10.9 ± 2.3 |
Cells were incubated with each drugs for 72 h. IC50s are showed as mean ± S.E.M. (n = 6–12). * p < 0.05, ** p < 0.01 compared with NRM or normal cell line.
In the SUIT-2 cells which had cultured with gemcitabine for long time, cell viabilities were significantly higher than those in normal SUIT-2 cells when cells were exposed to gemcitabine (Fig. 2A). IC50 values for gemcitabine was 4.67-fold higher in the cell lines which had cultured with gemcitabine for long time than in normal SUIT-2 cells (Table 1). Conversely, some statins and bisphosphonates were associated with slightly higher cell viabilities in the cells chronic-cultured with gemcitabine than in normal SUIT-2 cells (Figs. 2E–H, J, K). IC50 values for statins were 1.27–2.29-fold higher in the cells chronic-cultured with gemcitabine than in normal SUIT-2 cells (Table 1). These increases in IC50 values were less pronounced than those induced by gemcitabine. Furthermore, no significant differences in IC50 values of bisphosphonates were observed between the normal SUIT-2 cells and those chronic-cultured with gemcitabine (Table 1). Finally, no significant differences in IC50 values of statins or bisphosphonates were observed between the normal MIA PaCa-2 cells and those chronic-cultured with gemcitabine, although IC50 values of gemcitabine were significantly higher in the MIA-PaCa2 cells long-cultured with gemcitabine than in the normal cells (Table 1).
Cells were incubated with gemcitabine (A), oxaliplatin (B), paclitaxel (C), simvastatin (D), atorvastatin (E), rosuvastatin (F), fluvastatin (G), pitavastatin (H), pravastatin (I), zoledronate (J), and alendronate (K) for 72 h. Cell viabilities were measured using the WST-8 method. Open and closed circles mean the cell viabilities of normal cell line and those chronic-cultured with gemcitabine, respectively. Results are expressed as mean ± S.E.M. (n = 6–12). * p < 0.05, ** p < 0.01 compared with normal cell line.
In this study, we generated conditions under which pancreatic cancer cells could show poor responses to gemcitabine through long-term culture in either NDM or gemcitabine-containing medium, consistent with previous reports.1–4) Such conditions are believed to mimic chemotherapy resistance acquired in vivo within nutrient-deprived conditions, such as in the interior of large tumors, or after long-term gemcitabine therapy.
In nutrient-deprived conditions, high sensitivity to statins and bisphosphonates was displayed by both SUIT-2 and MIA PaCa-2 cells. In addition, the anti-cancer effects of statins and bisphosphonates in MIA PaCa-2 cells, which had been cultured long-term in gemcitabine-containing medium, were nearly equal to those observed in normal MIA PaCa-2 cells. Although IC50 values of statins were mildly elevated in SUIT-2 cells which cultured with gemcitabine chronically, these elevations were minuscule compared with the increase in IC50 values of gemcitabine. Also, SUIT-2 did not develop resistance to bisphosphonates in long-term culture with gemcitabine. These results suggest that HMG-CoA reductase inhibitors and bisphosphonates demonstrate anti-cancer activities in the pancreatic cancer cell lines which show poor responses to gemcitabine.
The mechanisms why nutrient-deprived condition induces the chemotherapy resistance remain unclear. Saiyin and colleague had reported that BR serine/threonine-protein kinase 2 (BRSK2) was induced by nutrient deprivation in pancreatic cancer cells and suppressed mammalian target of rapamycin C1 (TORC1) activity via phosphorylation of tuberous sclerosis complex 2 (TSC2).7) These mechanisms may play roles in the resistance to chemotherapy. The other studies had reported Hypoxia inducible factor 1 (HIF-1) is also related to the resistance in hypoxic and nutrient-deprived conditions.8,9) The lack of equilibrative nucleoside transporter 1 (ENT1) and deoxycytidine kinase (dCK) and increase in the levels of ribonucleotide reductase subunits M1 (RRM1) and M2 (RRM2) are involved in the development of gemcitabine resistance.10–13) Furthermore, a clinical study reported that the expression of these four key genes influences the prognosis of patients with pancreatic cancer undergoing adjuvant chemotherapies.14) ENT1 is a cellular transporter involved in the intracellular incorporation of gemcitabine. In contrast, lipophilic statins can penetrate the cell membranes via passive diffusion.15) Therefore, it is considered that statins provide reliable efficacy against pancreatic cancer cells even if the cells downregulate the expression of ENT1 in the process of acquiring gemcitabine resistance. The remaining factors—dCK, RRM1, and RRM2—are all specific to the action mechanisms and metabolic pathways of gemcitabine. Our institution previously reported that statins and bisphosphonates induce anti-proliferative effects via the downregulation of prenylation of RhoA and Ras in the mevalonate pathway in normal pancreatic cancer cell lines.5) Because these pathways are independent of gemcitabine-resistance related factors (dCK, RRM1, and RRM2), statins and bisphosphonates display anti-cancer effects also against the cell lines which show poor responses to gemcitabine. Furthermore, RhoA and Ras are not specific to pancreatic cancers. Therefore, HMG-CoA reductase inhibitors and bisphosphonates might be effective for other cancers.
In this study, we had not performed any experiments about the effects of combinations of HMG-CoA reductase inhibitors and the other drugs because we wanted to check the effects of these drugs when used alone respectively. Previous studies had reported the additive effects of HMG-CoA reductase inhibitors, bisphosphonates and gemcitabine for pancreatic cancer in cultured cells and clinical study.5,16) Thus, it is possible that combinations of HMG-CoA reductase inhibitors, bisphosphonates and gemcitabine show addictive effects for pancreatic cancer cell line which show poor responses to gemcitabine.
In conclusion, our results suggest that HMG-CoA reductase inhibitors and bisphosphonates have anti-cancer activities against pancreatic cancer cells which show poor responses to gemcitabine, established through long-term culture in either nutrient-deprived or gemcitabine-containing conditions. These effects were generally equitable to, or greater than, those observed in normal cancer cells. Recently, a meta-analysis has also indicated that statins improve the survival time of patients with pancreatic cancer.17) Therefore, statins and bisphosphonates may be effective in the treatment of pancreatic cancer, particularly in chemotherapy-resistant cases.
The authors declare no conflict of interest.
