Abstract
Background: Spectrin αII contributes to cisplatin and carboplatin resistance in ovarian serous carcinoma cells, and its expression in surgical specimens is a valid predictor of prognosis. We sought to identify effective drugs for spectrin αII-mediated cisplatin-resistant cells. Materials and Methods: We employed SKOV3 cells with small interfering RNA-mediated spectrin αII downregulation, serous carcinoma cells (NOS2), cisplatin-resistant cells (NOS2CR2), and oxaliplatin-resistant cells (NOS2OXR). Results: In the drug-sensitivity test, oxaliplatin was not affected by the inhibition of spectrin αII expression and was effective for cisplatin-resistant NOS2CR2 cells. NOS2OXR cells did not express higher levels of spectrin αII compared to NOS2 in western blot analysis. Six non-platinum anticancer drugs were not affected by the inhibition and was effective for resistant NOS2CR2 and NOS2OXR cells. Doxorubicin exhibited potent cytotoxicity at 2 μM against both resistant cell lines. Conclusion: Pegylated liposomal doxorubicin/oxaliplatin regimen may be effective for cisplatin-resistant ovarian carcinoma with spectrin αII-overexpression.
The most common course of treatment for ovarian carcinoma includes surgical intervention and chemotherapy (1). Many of the most effective chemotherapeutic drugs currently being used exhibit cytotoxic properties. Whereas several cytotoxic anticancer compounds are used for ovarian carcinoma treatment, paclitaxel and carboplatin combination chemotherapy (TC) is the first-line chemotherapy regimen used for ovarian carcinoma (2, 3). Although TC is highly effective against ovarian carcinoma, many advanced or clear-cell carcinomas do not respond well to this regimen. Recently, anticancer drugs, including bevacizumab (4, 5), and olaparib (6, 7), which have specific molecular targets, have been introduced into clinical therapy for the treating advanced or recurrent cases. Furthermore, immune checkpoint inhibitors are being developed as anticancer agents; however, their establishment as part of the standard treatment regimens will take time (8). Until then, TC will continue to be administered as a standard for chemotherapy, necessitating research on resistance to TC in order to overcome refractory ovarian carcinoma.
Not being able to determine the effectiveness of TC before treatment is problematic. TC is administered to all patients requiring chemotherapy. In patients with low levels of sensitivity, the carcinoma can proliferate despite anticancer drug treatment. Recurrent carcinoma cells develop resistance faster than wild-type cells after TC administration, and malignancies consequently relapse. The prognosis would improve if clinicians were able to identify whether a carcinoma is TC resistant before the administration of anticancer drugs, enabling the selection of appropriate medicines with the highest efficacy. Successful initial treatment regimens would reduce the rates of recurrence and improve survival rates.
A study using affinity chromatography with cisplatin-exposed glutathione sepharose 4B reported on the purification of the approximately 300-kDa cytoskeleton proteins spectrin αII and βII from serous ovarian cisplatin-resistant carcinoma cells (NOS2CR2) (9). NOS2CR2 cells exhibited higher expression of spectrin αII and βII in compared to the wild-type NOS2 cells. Moreover, sensitivity to cisplatin and carboplatin was found to be improved in cells expressing a low level of spectrin αII following transfection with small interfering RNA (siRNA). The researchers concluded that cytoskeletal spectrin αII and βII tetramers generate spectrin–glutathione–platinum complexes, thereby arresting cisplatin activity by anchoring the glutathione–platinum complexes to the membrane. They next performed a clinical retrospective study and demonstrated that cases of ovarian carcinoma positive for spectrin αII expression exhibited higher overall mortality rates and higher mortality following tumour recurrence compared to cases that were non-positive (negative or equivocal) for spectrin αII expression. Therefore, the positive expression of spectrin αII in surgical specimens is a useful predictor of anticancer drug resistance and postoperative prognosis (10).
Here, we sought to identify effective anticancer drugs for spectrin αII-mediated cisplatin-resistant cells, which might be clinically effective for refractory ovarian carcinoma. To ensure rapid clinical adoption if proven effective, we chose drugs that are already being used in daily clinical therapy. The identification of effective drugs would provide alternative treatment options for patients testing positive for spectrin αII in surgical specimens.
Materials and Methods
Cell culture and cell characteristics. Ovarian carcinoma cells (SKOV3), serous ovarian carcinoma cells (NOS2), and derived cisplatin-resistant cells (NOS2CR2) were prepared as previously described (9). SKOV3 cells were obtained from the American Type Culture Collection (ATCC®HTB-77; Manassas, VA, USA). NOS2 cells were established in our laboratory in 1986. Oxaliplatin-resistant cells (NOS2OXR) were developed by step-wise selection from NOS2 cells using oxaliplatin at a final concentration of 10 μM.
Analysis of spectrin αII expression. Western blot analysis of spectrin αII expression in NOS2 and NOS2OXR cells was performed using a primary antibody to spectrin αII (H-105) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and an internal control. β-Actin expression was assessed using antibody to β-actin (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan). Following staining with both primary antibodies, an anti-mouse antibody (Nichirei Histofine, Nichirei Bioscience Co. Ltd., Tokyo, Japan) was used as the secondary antibody at a 1:10 dilution (9). Immunohistochemical staining of NOS2 and NOS2CR2 cells was performed using SPTAN1/Alpha II-spectrin IHC antibody (Bethyl Laboratory, Inc., Montgomery, TX, USA) at a 1:100 dilution. The slides were developed using 3,3’-diaminobenzidine and were counterstained with methylene blue. The detailed methods of western blot analysis and immunohistochemical staining procedure were as described previously (9, 10).
Anticancer drugs. Cisplatin (Nichi-Iko Pharmaceutical Co., Ltd., Toyama, Japan), carboplatin (Nippon Kayaku Co., Ltd., Tokyo, Japan), nedaplatin (LKT Laboratories. Inc., St. Paul, MN, USA), and oxaliplatin (Sawai Pharmaceutical Co., Ltd., Osaka, Japan) were chosen as platinum anticancer drugs. Paclitaxel (Nippon Kayaku), vinorelbine (Nippon Kayaku), SN-38 (a bioactive metabolite of irinotecan: 7-ethyl-10-hydroxycamptothecin) (Merck KGaA, Darmstadt, Germany), doxorubicin (Nippon Kayaku), 5-fluorouracil (Kyowa Hakko Kirin Co., Ltd., Tokyo, Japan), and gemcitabine (Sandoz Co., Ltd., Ueyama, Yamagata, Japan) were employed as non-platinum anticancer drugs. These cytotoxic anticancer drugs except nedaplatin are described in the National Comprehensive Cancer Network Guidelines for patients, Ovarian Cancer 2019 (1). Nedaplatin is generally used in clinical treatment in Japan. Administered concentrations of each drugs are given in Table I.
siRNA transfection in SKOV3 cells. An anticancer drug sensitivity test was performed using SKOV3 cells in which spectrin αII expression was suppressed via siRNA transfection at 9 μg/ml. Transfection was performed using LipofectamineTM RNAiMAX Regent Invitrogen® (Thermo Fisher Science Co., Ltd., Tokyo, Japan) according to the manufacturer's instructions. Spectrin αII siRNA was designed by Nippon EGT Co., Ltd. (Toyama, Japan). The sense and antisense strands of spectrin αII siRNA-a (sense, UCAGUUGGAUUAAGGAAAATT; antisense, uuuuccuuaauccaacugaTT), siRNA-b (sense, GGAAAAGAUUACAGCAUUATT; antisense, uaaugcuguaauc uuuuccTT), and scrambled siRNA-c (sense, GGAAAGGUAUUUAAGUCAATT; antisense, uugacuuaaauaccuuuccTT) were reconstituted to a final concentration of 20 μM. Western blot analysis in this study confirmed the suppression of spectrin αII expression, which was enabled by the new transfection reagent employed in this study compared to a previous study (9).
Schedule of siRNA transfection and drug sensitivity measurement. On day 1, SKOV3 cells were seeded in a 24-well plate (Nunclon™ ΔSurface, Thermo Scientific, Roskilde, Denmark) at 1×105/well. On day 2, cells were transfected with LipofectamineTM RNAiMAX Regent Invitrogen® (Thermo Fisher Science Co., Ltd., Tokyo, Japan) according to the manufacturer's instructions. On day 3, the cells were washed twice with 10% foetal bovine serum and medium to remove siRNA and LipofectamineTM RNAiMAX from the cultures. The cultures were continued in culture medium alone. On day 4, the cells were transferred to a 96-well microplate (Thermo Scientific) at 3,000/well. Six wells were used for each concentration of the anticancer drugs. The cultures were incubated for 4 days. On day 8, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulphophenyl)-2H-tetrazolium (MTS) (CellTiter96® AQUEOUS One Solution Cell Proliferation Assay, Promega, Madison, WI, USA) was added, and absorbance was measured at 490 nm after 3 or 6 h incubation. Half-maximum inhibitory concentration (IC50) values were determined after repeating the measurement at least thrice for each culture.
Anticancer drug sensitivity assay. NOS2, SKOV3, NOS2CR2, and NOS2OXR cells were cultured in 96-well plates at 2,000, 3,000, 4,000, and 5,000 cells/well, respectively. Drug sensitivity was determined using the MTS assay. Cells were treated with the MTS reagent for 3 h (SKOV3, NOS2, and NOS2CR2 cells) or 6 hours (NOS2OXR cells). After the measurement was repeated at least thrice for each culture, sensitivity curves were created. Detailed methods of the anticancer drug sensitivity tests are available in a published study (9).
Results
Effect of spectrin αII suppression on sensitivity to nedaplatin and oxaliplatin. To identify effective anticancer drugs capable of overcoming platinum resistance in cells expressing spectrin αII, we suppressed spectrin αII expression in carcinoma cells (SKOV3). We confirmed that spectrin αII expression was down-regulated in the cells transfected with siRNA-a and siRNA-b compared to those that were not treated with siRNA, or were treated with scrambled siRNA-c (negative control). Because we used a different transfection reagent from that used earlier (9), we confirmed suppressive effects with the new reagent in the current study. The suppression effect in this study was similar to that in the previous study (Figure 1A).
The effects of spectrin αII on the sensitivity of cells to nedaplatin and oxaliplatin were examined. SKOV3 cells transfected with siRNA-a and siRNA-b showed increased sensitivity to nedaplatin in comparison with the negative control siRNA-c (Figure 1B). The sensitivity of cells to oxaliplatin was not affected by inhibition of spectrin αII (Figure 1C). The results showed that the IC50 of nedaplatin was reduced in SKOV3 cells transfected with either siRNA-a (61%, 1.85 μM) or siRNA-b (88%, 2.68 μM) compared to the negative control, siRNA-c (100%, 3.05 μM). The IC50 of oxaliplatin was not affected following transfection with either siRNA-a (86%, 3.83 μM) or siRNA-b (97%, 4.35 μM) compared to the negative control siRNA-c (100%, 4.46 μM) (Table II).
Sensitivity of cisplatin-resistant cells overexpressing spectrin αII to four platinum anticancer drugs. Spectrin αII overexpression has been determined via western blot analysis before (9); thus, we confirmed its expression using immunohistochemistry. Spectrin αII expression was observed at the cell membrane of cisplatin-resistant NOS2CR2 cells. However, minimal expression was observed in wild-type NOS2 cells, (Figure 2A). Sensitivity curves for the drugs are shown in Figure 2B–E. The IC50 and resistance ratios calculated for NOS2 and NOS2CR2 cells are shown in Table III. The resistance ratio was defined as the IC50 value of resistant cells/IC50 value of wild-type cells. The resistance ratios for cisplatin, carboplatin, and nedaplatin were calculated to be 67.1, 28.7, and 64.1, respectively. However, the resistance ratio for oxaliplatin was determined to be only 3.41.
Characteristics of oxaliplatin-resistant cells. Oxaliplatin sensitivity curves of NOS2 and NOS2OXR cells are shown in Figure 3A. The IC50 values for oxaliplatin were 1.19 μM in NOS2 cells, and 49.3 μM in NOS2OXR cells, indicating a 41.4-fold change in resistant cells (Table III). We also examined the expression of spectrin αII via western blot analysis (Figure 3B). Our results show that the expression of spectrin αII in NOS2OXR cells was similar to that observed in NOS2 cells. The sensitivity curves of NOS2 and NOS2OXR cells for cisplatin are shown in Figure 3C. The IC50 values for cisplatin in NOS2 and NOS2OXR cells were 1.48 and 16.3 μM, respectively, indicating an 11.0-fold change in resistant cells (Table III). Thus, cisplatin was weakly effective for oxaliplatin-resistant cells.
Effect of spectrin αII suppression on the sensitivity to non-platinum anticancer drugs. Previously, MTS assays showed that the IC50 of paclitaxel was not affected by reducing spectrin αII expression (9). Similarly, our MST assays showed that in SKOV3 cells transfected with siRNAs, sensitivities to vinorelbine, SN-38, doxorubicin, 5-fluorouracil, and gemcitabine did not change following inhibition of spectrin αII expression (Table II).
Sensitivity of cisplatin-resistant cells overexpressing spectrin αII to non-platinum anticancer drugs. Sensitivity curves for six non-platinum drugs are shown in Figure 4. The IC50 values and resistance ratios are presented in Table III. The resistance ratios for all six drugs were determined to be lower than that for cisplatin, which showed a 67.1. However, the graphs show cell proliferation of 15.1% following treatment with 100 nM paclitaxel (Figure 4A), 12.1% with 100 nM vinorelbine (Figure 4B), 21.1% with 1,000 nM SN-38 (Figure 4C), and 27.2% with 200 nM gemcitabine (Figure 4F). The graph of doxorubicin shows a cell proliferation of 4.6 % at 2 μM (Figure 4D). Although the sensitivity curves for 5-fluorouracil show a lower IC50 value for NOS2CR2 cells compared to NOS2 cells, the cell proliferation curves crossed over following treatment with 20 μM and 50 μM, resulting in a higher proportion of NOS2CR2 cells surviving compared to NOS2 cells when treated with 5-fluorouracil at 50, 100, and 200 μM. The graphs show a cell proliferation of 9.3% following treatment with 200 μM of 5-fluorouracil (Figure 4E). Thus, doxorubicin and 5-fluorouracil both exhibited strong cytotoxic effects on NOS2CR2 cells at high concentrations, resulting in <10% cell proliferation at the highest concentrations (Figure 4D and E), whereas NOS2CR2 proliferation was >10% for the other four non-platinum anticancer drugs.
Sensitivity of oxaliplatin-resistant cells to non-platinum anticancer drugs. As oxaliplatin was effective for spectrin αII-mediated cisplatin-resistant cells, we sought to identify a non-platinum anticancer drug to compensate for the weakness of oxaliplatin treatment for clinical applications. The sensitivity curves for the six non-platinum anticancer drugs are shown in Figure 5. The IC50 values and resistance ratios are shown in Table III. Resistance ratios for all six drugs were lower than the 41.4 resistance ratio for oxaliplatin. However, the five drugs, namely paclitaxel, vinorelbine, SN-38, 5-fluorouracil, and gemcitabine, had higher resistance ratios in NOS2OXR cells than in NOS2CR2 cells. The resistance ratio for doxorubicin in NOS2OXR cells was slightly lower than that observed in NOS2CR2 cells. Graphs revealed cell proliferation ratios of 41.1% following treatment with 100 nM paclitaxel (Figure 5A), 46.7% with 20 nM vinorelbine (Figure 5B), 16.6% with 5,000 nM SN-38 (Figure 5C), 39.1% with 200 μM 5-fluorouracil (Figure 5E), and 48.9% with 200 nM gemcitabine (Figure 5F). However, at a concentration of 2 μM, doxorubicin caused NOS2OXR cell proliferation to decrease to 6.3% (Figure 5D). The other five non-platinum anticancer drugs did not exhibit such high levels of cytotoxicity even at the highest concentrations tested. Therefore, doxorubicin was found to be the most effective of the non-platinum drugs in oxaliplatin-resistant cells.
Discussion
Spectrin αII contributes to cisplatin resistance in serous ovarian carcinoma, and patients with ovarian carcinoma with positive expression of spectrin αII have overall poorer prognoses compared to non-positive ones (9, 10). Here, we aimed to identify anticancer drugs to overcome drug resistance with spectrin αII overexpression. All the studied drugs are currently in clinical use for ovarian carcinoma. We found that oxaliplatin sensitivity was not affected by the expression of spectrin αII, and oxaliplatin was effective for cisplatin-resistant cells. Oxaliplatin-resistant cells did not show spectrin αII overexpression. Oxaliplatin was shown to overcome platinum resistance without being affected by spectrin αII. Next, in order to compensate for the weakness of oxaliplatin effects, we sought to identify non-platinum drugs that might be combined with oxaliplatin. The sensitivity of cells to the six non-platinum anticancer drugs was also not affected by the expression of spectrin αII, and these drugs were effective for cisplatin-resistant cells and oxaliplatin-resistant cells. Doxorubicin was the most cytotoxic at high concentrations among the six drugs. In the literature, phase II trials of the pegylated liposomal doxorubicin (PLD)/oxaliplatin regimen have already been reported, showing it to be safe and promising for refractory ovarian carcinoma (11-14). We concluded that the PLD/oxaliplatin regimen might be an effective treatment for ovarian carcinoma cases with spectrin αII-positive expression.
We demonstrated that reducing spectrin αII expression increased the efficacy of nedaplatin in this study. This has been demonstrated for cisplatin and carboplatin before (9); however, the same effect was not observed for oxaliplatin. Unfortunately, we were unable to transfect NOS2CR2 cells with spectrin αII siRNAs. Furthermore, only oxaliplatin among the platinum drugs was effective for cisplatin-resistant NOS2CR2 cells overexpressing spectrin αII in resistance ratio from 67.1 to 3.41. To investigate the effect of spectrin αII for oxaliplatin resistance, we developed oxaliplatin-resistant cells showing a 41.4-fold resistance compared with parental cells. In a previous study, carboplatin-resistant cells (NOS2CBR) with 16.9-fold resistance ratio exhibited enhanced expression of spectrin αII (9). However, the oxaliplatin-resistant cells and wild-type cells expressed similar levels of spectrin αII. Moreover, the cisplatin resistance ratio in oxaliplatin-resistant cells compared with wild-type cells was 11.0-fold. These results indicate that both partial cross-resistance mechanisms and individual independent resistance mechanisms may exist in cisplatin-resistant cells and oxaliplatin-resistant cells. Acquired resistance to cisplatin has been shown to result from loss of DNA mismatch repair, and to be primarily due to defects in the hMLH1 subunit of the hMutLα complex. However, oxaliplatin adducts are not recognized by the mismatch repair complex (15-17). Similarly, we found that spectrin αII contributed to resistance mechanisms against cisplatin, carboplatin (9), and nedaplatin activity but not to the resistance mechanisms against oxaliplatin.
Our result further demonstrated that the sensitivity to the six non-platinum anticancer drugs was not affected by spectrin αII. Based on the IC50 value, the six non-platinum anticancer drugs were effective for cisplatin-resistant cells and oxaliplatin-resistant cells. However, cell cultures exhibited high cell viability following treatment with a high concentration of the anticancer drugs with the five drugs except doxorubicin. Although 5-fluorouracil was more effective for cisplatin-resistant cells overexpressing spectrin αII than for the wild-type cells, the drug was not effective against oxaliplatin-resistant cells at a high concentration. 5-Fluorouracil could not be expected to compensate for the weakness of oxaliplatin. Doxorubicin was seen to reduce the resistance ratios in both cisplatin-resistant cells and oxaliplatin-resistant cells and demonstrated strong cytotoxic effects when applied to cultures at the highest concentration, 2 μM. The advantage of PLD over doxorubicin lies in the enhanced tissue permeability and retention of the nanocarriers in the former. We therefore propose that PLD would be an effective therapeutic option to overcome platinum drug resistance.
Based on the results above, oxaliplatin and PLD might be an effective combination therapy for treating patients with poor prognoses due to expression of spectrin αII in surgical specimens. Oxaliplatin (1, 18-20) or PLD (21, 22) can been administered to patients with recurrent ovarian carcinoma as a single agent. Several reports have cited the efficacy for the combination therapy of PLD and oxaliplatin. Recchia et al. described a phase I study involving PLD and oxaliplatin as a salvage chemotherapy for advanced ovarian cancer (11). Nicolett et al. reported on a phase II study that had an objective response rate of 28.6% resistant cases and 42.9% cases with stable disease cases (12). Palma et al. described cases involving complete regression and stabilization of bone marrow suppression with no allergic reactions in patients with ovarian cancer treated with PLD/oxaliplatin combination treatment, who had experienced adverse effects when administered carboplatin/paclitaxel treatment (13). Salah-Eldin et al. further described a 31.5% overall response rate and a 73.7% clinical benefit gained by treatment with PLD/oxaliplatin for ovarian cancer resistant to taxane–platinum treatment (14). Combination therapy with PLD and oxaliplatin would be more potent and effective than single-agent treatment.
Further clinical trials will investigate whether PLD and oxaliplatin combination chemotherapy can improve the prognosis of patients testing positively for spectrin αII. Although disease in most spectrin αII-positive patients exhibits cisplatin or carboplatin resistance, clinicians continue to administer cisplatin or carboplatin to patients with expression of spectrin αII. Our novel approach involves assessing the spectrin αII expression level in surgical specimens before assignment of a treatment regimen. TC would then be given only to patients not testing positive for spectrin αII. Patients testing positive for spectrin αII would be prescribed combination chemotherapy involving PLD and oxaliplatin. Bevacizumab may also be added for specific cases as needed. Thus, classification based on spectrin αII expression might be a novel test that predicts cisplatin and carboplatin drug sensitivity. This method would, therefore, prevent the administration of ineffective anticancer drugs and would ensure rapid and effective treatment after surgical interventions.
Recently, there has been a focus on developing drugs with specific molecular targets and immune checkpoint inhibitors; however, we believe that cytotoxic anticancer drugs will continue to be the first choice for ovarian carcinoma treatment. New drugs will be combined with known cytotoxic anticancer drugs, and clinical researchers will identify ways to improve patient prognosis. The PLD/oxaliplatin regimen is likely to be an effective treatment for ovarian carcinoma cases with spectrin αII-positive expression.
Acknowledgements
The Authors thank Kazuko Matsuba (Department of Pathology and Laboratory Medicine, Nagoya University of Hospital) for excellent technical skill in performing immunohistochemical staining. We would like to thank Editage (www.editage.jp) for English language editing.
Footnotes
Authors' Contributions
Osamu Maeda conceived and designed the study; contributed to data collection, analysis, and interpretation; and drafted the article. Hiroaki Kajiyama interpreted the data. Kiyosumi Shibata and Shigeo Nakamura contributed to data acquisition, and Fumitaka Kikkawa substantively revised and provided final approval of the article for publication.
Conflicts of Interest
None declared.
Funding
This research did not receive any specific grant from funding agencies in public, commercial, or not-for-profit sections.
- Received March 28, 2020.
- Revision received April 8, 2020.
- Accepted April 9, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved