Ovarian carcinoma (OVCA) is one of the most lethal malignancies of the female genital tract. Since OVCA frequently remains clinically silent, the majority of patients with this disease have advanced intraperitoneal metastatic disease at diagnosis.1 Recently, aggressive cytoreductive surgery and platinum-taxan-based chemotherapy have been employed in an attempt to improve the survival rate in patients with OVCA.2, 3, 4 Although OVCA cells are generally sensitive to anticancer drugs, certain histological types, such as mucinous and clear cell adenocarcinoma or recurrent tumors, are resistant to most anticancer drugs. However, the regulatory mechanisms responsible for drug resistance and progression in this tumor remain unclear.
Aminopeptidase N (APN), a Zn2+-dependent ectopeptidase localized on the cell surface, is a transmembrane protein that cleaves N-terminal neutral amino acids of various peptides and proteins.5 APN was originally characterized as a T-cell differentiation antigen (CD13).6 APN/CD13 activates or inactivates bioactive peptides on the cell surface by cleaving them enzymatically and thereby regulating their availability to nearby cells. Recent reports have indicated that APN/CD13 has a variety of functions, including functions in inflammatory and immunological responses, cell cycle control and antigen processing and presentation, as well as metabolizing certain vasoactive peptides.7, 8 Additionally, via these effects, APN/CD13 is thought to play diverse tissue-specific roles in both catalytic and noncatalytic ways.9, 10
A number of studies have provided evidence indicating that APN/CD13 may also play a role in various aspect of tumor progression, including mitogenic activation, tumor invasion and cell adhesion.10, 11, 12, 13 Furthermore, a recent examination showed that APN/CD13 was involved in protection of leukemic cells against apoptosis.14 In general, apoptosis is the predominant mechanism of cytotoxicity induced by chemotherapeutic agents. The failure of cancer cells to activate apoptosis may lead to multidrug resistance.
In the current study, we investigated the role of APN/CD13 in the chemosensitivity of OVCA. We first examined the correlation between the expression levels of APN/CD13 and sensitivity to paclitaxel (PAC) in OVCA cells. Subsequently, to clarify the cellular functions of APN/CD13, we investigated a role of this molecule in the PAC-sensitivity of OVCA using APN/CD13 inhibitor or siRNA specific for APN/CD13. A possible application of this enzyme inhibition as a means of enhancing PAC chemosensitivity to OVCA is proposed.
Material and methods
We used 10 human OVCA cell lines (SKOV-3, HEY, ES-2, NOS1, NOS2, TAOV, NOS4, RMG-I, RMG-II and KOC-7C) in this study. SKOV-3, NOS2, TAOV, NOS4, RMG-I and RMG-II cells were cultured and maintained as described previously.15, 16 ES-2, HEY and KOC-7C cells were purchased from the American Type Culture Collection (ATCC). NOS1 cells were established in our institute. The cell lines were maintained in RPMI-1640 supplemented with 10% fetal calf serum (FCS) and penicillin–streptomycin at 37°C in a humidified atmosphere of 5% CO2.
Flow cytometric analysis
Fluorescence-activated cell sorting (FACS) was performed to quantify the expression levels of APN/CD13 on the cell surface of OVCA cells as follows. The cells were incubated with phycoerythrin-conjugated monoclonal antibody specific for APN/CD13 (Pharmingen, San Diego, CA) for 30 min at 4°C, and washed 3 times with PBS. FACS data were acquired using a FACS Calibur (Becton Dickinson, San Jose, CA), and analyzed using CELL Quest software (Becton Dickinson).
Tissue samples of OVCA were obtained with informed consent from patients who were surgically treated at Nagoya University Hospital. Immunohistochemical staining of APN/CD13 was performed using the avidin–biotin immunoperoxidase technique (Histofine SAB-PO kit, Nichirei, Tokyo, Japan) as described earlier.15 Endogenous peroxidase activity was blocked by incubation with 3% H2O2, and nonspecific immunoglobulin binding was blocked by incubation with 10% normal rabbit serum. As a first antibody for APN/CD13 staining, anti-APN/CD13 mAb (Novocastra, Newcastle, United Kingdom) was used at a dilution of 1:100.
Inhibition of APN/CD13 expression by small interfering RNA
We designed and purchased 2 different siRNA duplexes targeting APN/CD13, siRNA1 (sense, 5′-CACCUUGGACCAAAGUAAA-3′) and siRNA2 (sense, 5′- GAAAUGCCACACUGGUCAA-3′) from Qiagen (Tokyo, Japan). Nonspecific scrambled siRNA duplex (sense, 5′-CUGGAUUGUAGGAAGUACCTT-3′) with the same GC content as APN/CD13 siRNA was purchased from Takara (Tokyo, Japan). The siRNA was transfected into ES-2 cells at a final concentration of 80 nmol/l using GenePorter-2 (Genlatis, San Diego, CA) according to the manufacturer's protocol.
Paclitaxel chemosensitivity assay
Cells were seeded in triplicate in 96-well plates at a density of 5,000 cells in a volume of 200 μl of culture medium containing 10% FCS. After incubation for 24 hr at 37°C, the medium was replaced with fresh medium with or without various concentrations of PAC (Bristol Myers Squib, Tokyo, Japan) or bestatin (Nihon-Kayaku, Tokyo, Japan). After an additional 72 hr, cell viability was assayed using a modified tetrazolium salt 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay performed with a Cell Titer 96 Aqueous One Solution Cell Proliferation Assay kit (Promega, Tokyo, Japan) according to the manufacturer's instructions. Absorbance was measured at 492 nm using a microplate reader (Multiskan Bichromatic; Labsystems, Helsinki, Finland). We also performed another PAC sensitivity assay using siRNA for suppression of APN/CD13. Briefly, cells were seeded in triplicate in 96-well plates at a density of 3,000 cells in a volume of 200 μl of culture medium containing 10% FCS. After incubation for 24 hr at 37°C, the medium was replaced with fresh medium containing 5% FCS in the presence of siRNA duplex of APN/CD13 or nonspecific control siRNA duplex using GenePorter-2 (Genlatis, San Diego, CA) according to the manufacturer's protocol. After 12 hr, the medium was replaced with fresh medium containing 5% FCS and various concentrations of PAC. After an additional 60 hr, cell viability was assayed as described above. IC50 values indicate the concentrations resulting in a 50% reduction in growth as compared with control cell growth. Experiments were performed in triplicate.
Assay for apoptosis
We also examined the cellular apoptosis under the APN/CD13-suppressive condition using bestatin or the siRNA technique. To quantify the apoptosis, annexin V and propidium iodide (PI) staining was performed followed by FACS. The preparation of the cells is described in “Figure legends.” After treatment, both floating and attached cells were collected by brief trypsinization and washed with PBS twice, and then subjected to annexin V and PI staining using an MEBCYTO apoptosis kit (MBL). After staining, quantitative analysis of apoptosis was performed using a FACS Calibur (Becton Dickinson, San Jose, CA), and analyzed using CELL Quest software (Becton Dickinson).
Western blot analysis
Cell lysates were electrophoresed in sodium dodecyl sulfate polyacrylamide gels under reducing conditions. After electrophoresis, the proteins were transferred electrophoretically to Immobilon membrane (Millipore, Bedford, MA). After blocking, the membrane was incubated for 1 hr with the respective antihuman primary antibody at the recommended dilution (anti-APN/CD13; LAB VISION, anti-Apaf and CAS: BD Transduction Laboratories, anti-phospho-p53, total p53 and Bax: Cell Signaling, anti-phospho-ERK, ERK, p21/WAF1 and β-actin: Santa Cruz). The membrane was washed 3 times with Tween/PBS for 15 min, and then incubated with the appropriate secondary antibody for 1 hr. After washing with Tween/PBS, the membrane was treated with ECL-Western blotting detecting reagent (Amersham Biosciences K.K., Tokyo, Japan).
In vivo studies
Female nude mice (BALB/c) at 5 weeks of age were obtained from Japan SLC (Nagoya, Japan). The treatment protocol followed the guidelines for animal experimentation adopted by Nagoya University. ES-2 cells (1 × 107 cells per 0.5 ml of medium/mouse) were injected i.p. to generate peritoneal metastasis of OVCA in the mouse model. The survival time was examined with or without treatment with PAC and/or bestatin.
Mice were divided into 4 groups, Group A, PBS alone; Group B, PAC alone; Group C, bestatin alone; Group D, both PAC and bestatin. Each group consisted of 8 mice, which were randomized before the treatment was initiated. Intraperitoneal (i.p.) administration of PAC (20 mg/kg body weight), or bestatin (20 mg/kg body weight) was initiated 48 hr after tumor inoculation, and was repeated 2 (PAC) or 3 (bestatin) times a week, respectively, for at most 32 days. In the control group (n = 7), vehicle alone was administered in the same manner. Survival time was compared among these 4 groups.
In vivo siRNA treatment
A total of 2.0 × 106 ES-2 cells were subcutaneously inoculated in 0.2 ml of serum-free RPMI1640 medium through a 24-gauge needle into the lower flank of 5-week-old nude mice (Japan SLC, Nagoya, Japan). After 3 days, the mice were treated with APN/CD13-siRNA (n = 5) or scrambled-siRNA (n = 5) with atelocollagen (Koken, Tokyo, Japan). The final concentration of atelocollagen was 1.75%. On days 3, 11 and 19, APN/CD13- siRNA (10 μM) or scrambled- siRNA (10 μM) was mixed with atelocollagen, and 100 μl of each mixture were injected into the tumor region. Simultaneously, intraperitoneal administration of paclitaxel (20 mg/kg body weight) was performed together with this treatment, which was initiated 2 weeks after ES-2 cells-inoculation and repeated every 3 days for at most 4 times. Tumor volume was measured with calipers and calculated using the formula: π/6 × (larger diameter) × (smaller diameter)2. Data are presented as mean ± SE.
For data of in vitro experiments, statistical comparisons among groups were performed using Student's t-test and ANOVA with Bonferroni corrections. Differences between groups were considered statistically significant at p < 0.05. Data are expressed as mean ± SD. In addition, Pearson's correlation test was used to compare the IC50 values of PAC and the mean fluorescence intensities of APN/CD13 among various OVCA cell lines by FACS analysis. Moreover, the Kaplan–Meier method was used to generate survival curves, and comparisons were performed using logrank tests. p < 0.05 was considered significant.
APN/CD13 expression in OVCA tissues and cell lines
We first examined APN/CD13 expression in surgically resected OVCA tissues. APN/CD13 was generally expressed in tumor cells of OVCA tissue, although the intensity of immunohistochemical staining varied from tissue to tissue (Fig. 1).
We next examined APN/CD13 expression in OVCA cell lines by FACS analysis (Table I). SKOV-3, ES-2 and HEY cells were positive for APN/CD13 expression. On the other hand, NOS2, TAOV, NOS4 and RMG-I cells showed limited expression of APN/CD13 in this analysis, which is consistent with data obtained by enzyme activity analysis (data not shown).
Table I. Correlation Between APN/CD13 Expression and IC50 Values of Paclitaxel in Various Ovarian Carcinoma Cell Lines
Correlation between APN/CD13 expression and IC50 values of PAC in OVCA cells
Table I shows the chemosensitivity of various OVCA cell lines to PAC. The IC50 values of PAC in NOS2, TAOV, NOS4 and RMG-I cells were less than 10 ng/ml of PAC. In contrast, those of SKOV-3, ES-2 and HEY cells were 33 ± 11, 41 ± 4.2 and 43 ± 8.2 ng/ml, respectively. Figure 2 shows the correlation between mean fluorescence intensity (MFI) of APN/CD13 and IC50 value in 10 OVCA cell lines. The results showed that the MFI of APN/CD13 and IC50 of PAC were positively correlated (p < 0.05).
The effect of bestatin on sensitivity to PAC in APN/CD13-expressing OVCA cells in vitro
To investigate the role of APN/CD13 in the sensitivity to PAC in ES-2 cells that highly expressed APN/CD13 and were only weakly sensitive to PAC, we used bestatin, an APN/CD13 inhibitor, which has been widely used for the suppression of APN/CD13 activity.14, 17, 18, 19 We first confirmed that the enzyme activity of APN/CD13 was suppressed by bestatin in ES-2 cells (data not shown). We then investigated the effect of bestatin on PAC-sensitivity using this cell line. As shown in Figure 3a, treatment with bestatin alone (0–200 μg/ml) did not significantly influence the proliferative potential of ES-2 cells (panel a). In contrast, a significant growth-inhibitory effect of bestatin was found in ES-2 cells in the presence of various concentration of PAC (50, 100 and 200 ng/ml, shown in panels b–d, respectively). Figure 3b shows the PAC-sensitivity curve in the presence of 100 μg/ml of bestatin in ES-2 cells. Bestatin (100 μg/ml) enhanced the sensitivity to PAC by up to ∼40% compared with the control with PAC alone. Cell death with morphological changes associated with apoptosis was more frequently observed in ES-2 cells treated with both PAC and bestatin than in nontreated cells or cells treated with PAC alone (Fig. 3c).
To further investigate the mechanism of APN/CD13-induced chemoresistance, we also assessed whether bestatin had an effect on PAC-induced apoptosis of ES-2 cells. Cells were treated with PAC (100 ng/ml) and/or bestatin (100 μg/ml) for 60 hr and then subjected to annexin V/PI staining and flow cytometric analysis. As shown in Fig. 3d, ES-2 cells treated with both PAC and bestatin exhibited a synergistically stronger apoptotic effect than cells treated with PAC alone. The quantitative data showed that compared to cells treated with PAC alone (1.53% apoptotic cells), cells treated with both bestatin and PAC resulted in a significant increase in apoptosis (38.2% apoptotic cells). We next investigated whether p53 was involved in the action of APN/CD13. We also examined whether the expression of apoptosis markers was affected by concurrent treatment with PAC and bestatin using Western blot analysis. The expression levels of p21/WAF1, Bax and phosphorylated p53 were upregulated in ES-2 cells treated with both PAC (100 ng/ml) and bestatin (100 μg/ml) compared with those in ES-2 cells treated with PAC alone. On the other hand, the expression levels of CAS, Apaf and total p53 were not affected by this treatment (Fig. 3e).
Effect of inhibition of APN/CD13 using siRNA on PAC-sensitivity of ES-2 cells
To further confirm the role of APN/CD13 expression in the chemoresistance of OVCA, we examined the effect of APN/CD13 suppression using the siRNA system on PAC-sensitivity in ES-2 cells. Two siRNAs (siRNA1 and siRNA2) specific for suppressing the APN/CD13 protein level were developed. Since siRNA1 was more efficient for reducing of APN/CD13 expression (data not shown), we used siRNA1 in the subsequent experiments. We confirmed the down-regulation of APN/CD13 expression at 24, 48, 72 and 96 hr after siRNA-transfection. The APN/CD13 protein level was most efficiently decreased after 72 hr of transfection (Fig. 4a). Figure 4b shows that the enhancement of PAC-sensitivity was more evident in APN/CD13-inhibited cells (ES-siCD13 cells) than in nonspecific control siRNA duplex-transfected cells (ES-siN cells). The IC50 values of ES-siCD13 and ES-siN cells were 13 ± 3.2 and 38.2 ± 4.1, respectively.
We next investigated whether a MAPK isoform, extracellular signal regulated kinase (ERK), was involved in the action of APN/CD13. Western blotting with a phosphospecific antibody showed that the phosphorylation levels of ERK were downregulated by the suppression of APN/CD13 by siRNA under conditions of serum-stimulation (Fig. 4c). In contrast, the expression level of total ERK was not influenced by this treatment. Moreover, the quantitative apoptotic data showed that ES-siCD13 cells showed a marked increase in the induction of apoptosis (17.7% apoptotic cells) compared with ES-siN cells (3.8% apoptotic cells) by PAC-treatment (Fig. 4d).
The effect of bestatin on the peritoneal progression of OVCA cells in a nude mouse model
Finally, we tested whether bestatin influenced the formation of peritoneal dissemination and survival time in ovarian carcinoma in a model using nude mice. Carcinomatous peritonitis was observed for ∼2 weeks after the inoculation of ES-2 cells into the mice. Figure 5a shows the general appearance of the mice 30 days after inoculation of ES-2 cells and treatment with PAC (Group B) or with PAC-bestatin (Group D). Group D-mice had more disseminated tumors and a larger amount of bloody ascites compared with Group B-mice. Figure 5b shows the survival curves of mice in Groups A–D. The survival times of Groups A (no treatment) and C (bestatin alone-treatment) mice were not significantly different (mean survival times; 16.7 ± 3.6 and 15.1 ± 2.6 days, respectively). Thus, treatment with bestatin alone did not improve the survival of mice with peritoneal dissemination. However, the mean survival times of Group B and D mice were 27.1 ± 6.6 and 37.7 ± 7.0 days, respectively. Thus, the survival time of Group D mice was significantly longer than that of Group B mice (p < 0.05). This result indicates that combined treatment with bestatin and PAC improved the survival time of mice with carcinomatous peritonitis.
Treatment of established ES-2 xenograft with APN/CD13 siRNA
ES-2 cells (2.0 × 106) were injected s.c. into the flank of nude mice. To examine the therapeutic effectiveness of APN/CD13-siRNA, intratumoral treatment with APN/CD13-siRNA with atelocollagen or scrambled-siRNA with atelocollagen was performed, accompanied by paclitaxel-treatment as described in “Material and methods.” Tumor growth curves are shown in Figure 6. APN/CD13-siRNA significantly suppressed tumor growth compared with scrambled-siRNA on days 20, 22 and 24 (p < 0.05).
PAC is a first-line chemotherapeutic agent that is effective for the treatment of OVCA. PAC exerts its effect through stabilization of microtubules, induction of cell cycle arrest in G2-M and activation of proapoptotic signaling.20, 21 According to previous reports, PAC treatment causes the induction of apoptotic pathways, activation of p53, Cyclin-dependent kinases, phosphorylation of Bcl-2 and the c-Jun NH2-terminal kinase/stress-activated protein kinase signaling pathway.22, 23, 24 However, in spite of the comparatively high sensitivity of OVCA to PAC, the prognosis of advanced or recurrent cases remains poor since most deaths are the result of metastasis that is refractory to conventional chemotherapy. To overcome the PAC-resistance, a variety of additional molecular-targeting therapies combined with PAC have been investigated.25, 26
In the current study, we first focused on bestatin as a drug used in combination with PAC. Bestatin is a dipeptide immunostimulator that was first isolated from a culture filtrate of Streptomyces olivoreticuli and that enhances the concanavalin A-induced activation of lymphocytes.27, 28 Bestatin inhibits aminopeptidase N, aminopeptidase B and leucine aminopeptidase of mammalian cells.29 An antigrowth or antimigratory effect is frequently observed in studies using bestatin or anticatalytic antibodies for APN/CD13 in several malignancies.10, 30, 31, 32
Previous studies have shown that bestatin induces an enhancement of apoptosis or increased sensitivity to antineoplastic agents. Mishima et al. showed that the increased IL-8 production from vascular endothelial cells caused by bestatin-treatment resulted in enhanced apoptosis of leukemic cells. In addition, Hirano et al. demonstrated that bestatin enhanced the sensitivity of acute promyelocytic leukemia cells to all-trans retinoic acid.33 Consistent with those findings, the present data suggested that bestatin enhanced the chemosensitivity of solid tumors to PAC.
In our animal model, treatment with bestatin alone did not influence the survival of nude mice inoculated i.p. with ES-2 cells, which are highly metastatic OVCA cells. However, combined treatment with bestatin plus PAC resulted in significantly prolonged survival, compared with treatment with PAC alone. These results are consistent with in vitro data shown in Figure 3a. This is the first report showing a link between APN/CD13 activity and chemosensitivity to PAC in OVCA in vitro and in vivo. The mechanism by which APN/CD13 activity functions in the regulation of PAC sensitivity remains to be clarified. APN/CD13 has been suggested to be involved in the degradation of neuropeptides, cytokines and immunomodulatory peptides, as well as angiotensins.34, 35 Thus, it is possible that APN/CD13 activity may contribute to chemoresistance through proteolytically modifying peptides involved in antiapoptotic signaling, and/or their precursors.
In the present study, expression of APN/CD13 and PAC-sensitivity were negatively correlated in various OVCA cell lines. Clinically, in pancreatic or colon cancer, the survival rate of patients negative for APN/CD13 expression is better than that of patients positive for APN/CD13 expression.31, 36 Indeed, bestatin affects the enzymatic activity of APN/CD13. However, in addition to affecting APN/CD13 activity, it is possible that APN/CD13 may also promote PAC-resistance by aminopeptidase-independent mechanisms. To suppress all of the functions of APN/CD13, an siRNA technique was used in our current PAC-sensitivity assay. The results revealed that the suppression of APN/CD13 expression by siRNA-induced enhanced chemosensitivity to PAC in OVCA cells. van Hensbergen et al. reported that APN/CD13-overexpressing ovarian cancer cells are less sensitive to cisplatin in a subcutaneous xenograft model using nude mice.37 Our current findings are consistent with that report. Cisplatin is an important chemotherapeutic agent for treatment of solid tumors, including ovarian carcinoma. Although cisplatin-sensitivity was not assessed in our current investigation, APN/CD13 might be involved in chemoresistance to both of these chemotherapeutic agents.
In this study, the apoptotic effect in PAC/bestatin-treatment was somewhat stronger than that in PAC/siRNA-treatment. One of the possible explanations for this observation is that bestatin may inhibit other aminopeptidase activities and thereby influence cell cycle or apoptotic signals since it is not completely specific for APN/CD13. Santos et al. demonstrated that APN/CD13 was directly involved in signal transduction pathways, including the phosphorylation of the mitogen-activated protein kinases, such as ERK1/2, JNK and p38, in monocytes.9 Consistently, our result also showed that suppression of APN/CD13 using siRNA resulted in decreased phophorylation of ERK1/2. Although the detailed mechanisms are still under investigation, APN/CD13 may be involved in signaling cascades regulating cell survival, protection from apoptosis and PAC-resistance via aminopeptidase-dependent and independent mechanisms.
In conclusion, the data obtained here suggest that the expression of APN/CD13 is linked with PAC-resistance, and inhibition of APN/CD13 may lead to the restoration of PAC-sensitivity in OVCA cells. Ichinose et al. showed that bestatin had a therapeutic effect of prolonging survival when used as a postoperative adjuvant treatment for patients whose stage I squamous-cell carcinoma was completely resected.38 Likewise, APN/CD13-targeting treatment such as bestatin combined with paclitaxel-based chemotherapy may lead to better prognosis of OVCA patients. Although further clarification of the mechanism and function of APN/CD13 in tumor biology is essential, APN/CD13 may become a novel therapeutic target for OVCA.