Skip to main content
NCBI home page
Neoplasia (New York, N.Y.) logoLink to Neoplasia (New York, N.Y.)
. 2008 Sep;10(9):964–972. doi: 10.1593/neo.08604

Clusterin Interacts with Paclitaxel and Confer Paclitaxel Resistance in Ovarian Cancer1,2

Dong Choon Park *,†, Seung Geun Yeo ‡, Mark R Wilson §, Justin J Yerbury §, Joseph Kwong †, William R Welch , Yang Kyu Choi #, Michael J Birrer **, Samuel C Mok ††, Kwong-Kwok Wong ††
PMCID: PMC2517641  PMID: 18714397

Abstract

Optimal debulking followed by chemotherapy is the standard treatment of managing late-stage ovarian cancer, but chemoresistance is still a major problem. In this study, we compared expression profiles of primary tumor tissue from five long-term (>8 years) and five short-term (<2 years) ovarian cancer survivors and identified clusterin as one of the genes that were significantly up-regulated in short-term survivors. We then evaluated the prognostic significance of clusterin and its possible correlation with chemoresistance in ovarian cancer by immunohistostaining of clusterin in 62 tumor samples from patients with stage III, high-grade serous ovarian cancer. After adjusting for debulking status and age, Cox regression analyses showed that high levels of clusterin expression correlate with poor survival (hazard ratio, 1.07; 95% confidence interval, 1.002–1.443; P = .04). We also investigated clusterin in paclitaxel resistance by modulating the endogenous clusterin expression in ovarian cancer cells and treating the cells with purified clusterin. Results indicate that high-clusterin-expressing ovarian cancer cells are more resistant to paclitaxel. Moreover, exposing ovarian cancer cells to exogenous clusterin increases cells' resistance to paclitaxel. Finally, using size exclusion chromatography and fluorescently labeled paclitaxel, we demonstrated that clusterin binds to paclitaxel. In summary, our findings suggest that high levels of clusterin expression increase paclitaxel resistance in ovarian cancer cells by physically binding to paclitaxel, which may prevent paclitaxel from interacting with microtubules to induce apoptosis. Thus, clusterin is a potential therapeutic target for enhancing chemoresponsiveness in patients with a high-level clusterin expression.

Introduction

Epithelial ovarian cancer, the most common human ovarian cancer [1], is also the most lethal, partly because 75% of ovarian cancers are detected as late-stage disease [2] and 25% of tumors do not respond to standard chemotherapy. Nevertheless, after surgical debulking of the tumor and standard chemotherapy, approximately 5% to 10% patients experience 5-year progression-free survival and are therefore considered long-term survivors [3]. Identifying new molecular therapeutic targets for the treatment of the disease should improve these outcomes, but to do so, we must first gain greater insight into the pathogenesis of ovarian cancer, perhaps by better understanding the molecular differences between tumors from short- and long-term ovarian cancer survivors.

Transcription profiling, one increasingly standard approach to investigating global gene expression differences in cancer samples [4–10], has already been used to identify many potential prognostic biomarkers and therapeutic targets. In this study, our primary purpose was to identify a list of genes that are significantly up-regulated in short-term survivors by transcription profiling. It is hoped that some of these genes might be potential therapeutic targets. Clusterin was found to be one of the genes that are significantly up-regulated in short-term survivors.

Although previous studies have established that high levels of clusterin expression correlate with the progression of various cancers, such as cancer of the bladder [11], breast [12], and prostate [13], and that clusterin may also be involved in chemoresistance in these cancers [14–16], the correlation of high clusterin expression in ovarian cancer cells with short-term survival and chemoresistance in ovarian cancer patients has not been established. Therefore, in this study, we sought to clarify the prognostic significance of clusterin in ovarian cancer and explore its role in chemoresistance. By manipulating clusterin expression and exposure in several ovarian cancer cell lines and using size exclusion chromatography and fluorescently labeled paclitaxel, we found that clusterin increases the resistance of ovarian cancer cells to paclitaxel by binding to it, both in vivo and in vitro. Furthermore, immunohistochemical analyses of 62 independent ovarian tumor samples from patients with stage III, high-grade serous ovarian cancer showed that high levels of clusterin expression correlate with poor overall survival rates.

Materials and Methods

Microarray Analysis

Affymetrix U133plus2 GeneChips (Santa Clara, CA) were used to generate expression profiles of microdissected tumor cells from five long-term (survival time, 95–192 months) and five short-term (survival time, 13–21 months) survivors with stage III, grade 3 serous ovarian cancer as described previously (Figure 1) [17]. All patients received an optimal debulking surgery and treatment with six cycles of paclitaxel and cisplatin. Raw images (“.DAT” files) from an Affymetrix GeneChip scanner were processed with dChip software [18]. The raw signals of individual probes for all 10 arrays were normalized against the chip with the median raw signal intensity, and normalization was based on an “invariant set” of probes consisting of points from nondifferentially expressed genes. After normalization, the expression values of each gene in all the samples were computed using a perfect match-only model and an outlier detection algorithm. Normalized expression values from dChip analyses were used for a two-class unpaired SAM analysis. Differentially expressed genes were identified by supervised analysis with Significance Analysis of Microarrays (SAM) software [19], which yields results similar to those obtained using Student's t test but includes permutations to calculate the false discovery rate, or a q value, for each gene. Each q value represents the probability that a gene is falsely identified as differentially expressed, with smaller q values indicating more significant differential expression levels.

Figure 1.

Figure 1

Open in a new tab

Partial list of differentially expressed genes in tumors from long- and short-term ovarian cancer survivors. The length of survival for each patient is shown within a labeled bracket. Red, white, and blue indicate a fold-change expression level above, at, and below the mean expression of a gene across all samples, respectively.

Cell Lines and Tissue Samples

All 13 ovarian cancer cell lines (PEO4, CAOV3, OVCA420, TOV21G, RMUG-S, RMUG-L, OVCA822, 0V2008, ES2, TOV112, OVCA433, OVCA432, and SKOV3) were grown in a medium of equal parts M199 and MCDB105 supplemented with 10% fetal bovine serum as described previously [20]. All tumor specimens from patients were collected and archived according to protocols approved by the institutional review boards of the appropriate institutions that included an informed consent or a waiver. All the patients had stage III and grade III serous ovarian tumors and were treated uniformly with paclitaxel/carboplatin first-line chemotherapy regimen.

Immunohistochemistry

Five normal ovarian surface epithelial samples and 62 samples from previously untreated stage III, high-grade primary papillary serous carcinomas were immunostained using an avidin-biotin method with an anti.clusterin α chain mouse monoclonal antibody (5 µg/ml; Upstate Biotechnology, Lake Placid, NY), with clusterin expression scored as described previously [21]. Cox regression and Kaplan-Meier survival analyses were used to compare immunohistochemical results with patient survival data. Statistical significance was determined by using the log rank test with statistical software SPSS version 15.0.

Western Blot Analysis

Cell lysates were prepared from the 13 ovarian cancer cell lines [20] and subjected to Western blot analysis using an anti-clusterin mouse monoclonal antibody (Upstate Biotechnology) and an anti-β-actin monoclonal antibody for normalization of protein loading. Immunoreactivity was detected using the ECL chemiluminescence system (Amersham, Piscataway, NJ) and quantified using an imaging densitometer (Model GS-670; Bio-Rad, Hercules, CA).

IC50 and Cell Proliferation Assay

Cells (1 x 104 per well) were seeded in a 96-well plate in 0.1 ml of culture medium and allowed to grow overnight. The next day, the medium was supplemented with either vehicle control or 10-9 to 10-4 M paclitaxel (Sigma, St. Louis, MO). The 50% inhibitory drug concentration (IC50) was determined using the 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) assay (Roche Molecular Biochemicals, Indianapolis, IN) according to the manufacturer's instructions. The metabolism of XTT was quantified by measuring the absorbance at 450 nm, and the IC50 of paclitaxel for each of the cell lines was estimated from the semilogarithmic dose-response curves by linear interpolation.

DNA Fragmentation Assay for Apoptosis

Cells (1 x 104 per well) were seeded in a 96-well plate in 0.1 ml of culture medium and allowed to grow overnight. The next day, cells were treated with 10-7 M paclitaxel for 48 hours at 37°C in the presence or absence of clusterin. The amounts of cytoplasmic histone-associated DNA fragments (mononucleosome and oligonucleosomes) formed during apoptosis were measured using a cell death detection ELISAPlus kit (Roche Molecular Biochemicals) according to the manufacturer's instructions. Each assay was performed in quadruplicate, and each experiment was repeated three times. The relative percentages of DNA fragmentation were calculated as the ratios OD405 readings from paclitaxel-treated to paclitaxel-untreated cells.

Generation of Stably Transfected Cells Overexpressing Clusterin

SKOV3 cells were transfected with full-length clusterin cDNA cloned into pcDNA3.1 expression vectors (pcDNA3.1/CLU) using FuGENE 6 transfection reagent according to the manufacturer's protocol (Roche Molecular Biochemicals). Mock transfection was performed the same way, using the vector pcDNA3.1. Stably transfected clones were selected 36 hours later by adding G418 (Roche Molecular Biochemicals) at 500 µg/ml. The stable clones were then evaluated for the expression of clusterin and paclitaxel resistance.

In Vitro Small Interfering RNA-Mediated Gene Silencing

Clusterin small interfering RNA (siRNA) duplex, consisting of nucleotides +85 to +96 (where the translation start site was defined as +1; Dharmacon Research, Lafayette, CO), or luciferase GL2 Duplex (Dharmacon) at 0.2 nmol/ml was transfected into ovarian cancer cells expressing high levels of clusterin (PEO4, RMUG-S, and TOV21G) using an oligofectamine reagent (Invitrogen, Carlsbad, CA) as described by the manufacturer's protocols (Life Technologies, Inc., Gaithersburg, MD). Transfected cells were evaluated for clusterin expression and sensitivity to paclitaxel.

Exogenous Clusterin Treatment

Clusterin was purified from human sera by immunoaffinity chromatography, using an anti-clusterin monoclonal antibody under native conditions at 4°C as described previously [22]. Analysis of clusterin purity by SDS-PAGE showed a major band (nonreduced) at 75 to 80 kDa [22]. To further ensure the purity of the product, we performed an additional purification step using ion-exchange chromatography. The eluate from the immunoaffinity column was dialyzed against 50 mM Tris buffer (pH 8.0), loaded onto a Q-Sepharose FF column, and eluted with a linear NaCl gradient in the same buffer. Clusterin-containing fractions (as judged by SDS-PAGE) were then pooled and concentrated. SKOV3 cells (1 x 104 cells per well) in 96-well plates were treated with 10-7 M paclitaxel in the presence of 0.5, 2.5, 5, and 7.5 and 15 µg/ml purified clusterin for 48 hours at 37°C. Cells were then harvested for XTT and apoptosis assays.

In Vivo Short Hairpin RNA-Mediated Gene Silencing Using a Lentiviral Expression System

A Lentiviral vector expressing short hairpin RNA (shRNA) targeting clusterin was constructed using the BLOCK-iT Lentiviral RNAi expression system (Invitrogen). DNA oligonucleotides encoding shRNA of clusterin (CLU) were designed using BLOCK-iT RNAi designer (Invitrogen) and inserted downstream of the human U6 pol III promoter and upstream of a pol III terminator in pENTR/U6 vector (Invitrogen) and then into the vector pLenti-GW/U6 as pLenti-GW/U6-CLUshRNA. Lentiviral particles were then generated from pLenti-GW/U6-CLUshRNA by transfecting 293 FT cell lines with the shRNA construct and helper plasmids.

To investigate the effects of decreased clusterin expression on paclitaxel resistance in xenografted tumors, forty 6- to 8-week-old female nude mice (SLC; Shizuoka, Japan) were evenly distributed for generating xenograft mouse models using control cells (PEO4 transduced with Lentiviral particles carrying shRNA for the LacZ gene) or clusterin knockdown cells (PEO4 transduced with Lentiviral particles carrying shRNA for the CLU gene). After being transduced with Lentiviral particles carrying shRNA for the LacZ or CLU gene overnight, 5 x 105 paclitaxel-resistant PEO4 cells were used to inoculate subcutaneously into the posterior neck region of each nude mouse. When the xenografts grew to approximately 100 mm3, mice were injected intraperitoneally with 0.2 mg/0.1 ml per 10 g of paclitaxel three times per week for 3 to 4 weeks. Tumor volumes were measured twice every week and calculated using the formula: length x width x depth x 0.5236. The mice were euthanized at day 27, and the levels of clusterin expression in xenografted tumors were determined by Western blot analysis. Animals were housed in pathogen-free units in compliance with institutional animal care and use committee regulations.

Effects of Clusterin on the Binding of Paclitaxel to Microtubules in Ovarian Cancer Cells

PEO4 and SKOV3 cells were trypsinized, inoculated into chambered coverglasses (Fisher Scientific, Pittsburgh, PA), and incubated for 12 hours at 37°C. The cells were washed three times with 2% BSA/PBS, 10-6 M Oregon Green (OG) 488-labeled paclitaxel (paclitaxel-OG; Invitrogen) was added, and the cells were incubated for 1 hour at 37°C in PEM buffer (50 mM PIPES, 2 mM EGTA, 2 mM MgCl2, pH 7.4), with or without 7.5 mg/ml purified clusterin. After incubation with the 10-6 mg/ml Hoechst 33258 pentahydrate (Molecular Probes, Eugene, OR) for 15 minutes at 37°C, the cells were examined using a Leica DMIRE2 inverted fluorescence microscope with appropriate filters.

Measurement of Paclitaxel-Clusterin Binding

Solutions of clusterin or control protein glutathione-S-transferase (GST) at 5 mM, alone or in the presence of a 1:1 molar ratio of paclitaxel-OG, phalloidin-OG, or OG in PEM were incubated for 1 hour at 37°C. Then, solutions were fractionated using size exclusion chromatography with a Biosep SEC S4000 column (Phenomenex, Sydney, Australia) and an AKTA FPLC system (GE Healthcare, Sydney, Australia); the A280 of the eluate, as an indication of protein elution, was measured continuously. Collected fractions (200 ml each) were transferred into the wells of black 96-well plates (Greiner, Frickenhausen, Germany), and the fluorescence of paclitaxel-OG, phalloidin-OG, and OG was measured using a Fluostar microplate reader (BMG Labtech, Melbourne; excitation, 485 ± 5 nm; emission, 520 ± 5 nm). Other control proteins (superoxide dismutase and hemoglobin) were also tested.

Results

Expression Profiling of Microdissected Tumors from Short- and Long-term Survivors

Using SAM analysis to compare the expression profiles of tumors from five short- and five long-term ovarian cancer survivors, we identified 77 probe sets with greater than threefold expression in shortthan in long-term survivors (Table W1; q < 15%). Two of these probe sets encode clusterin that was up-regulated 3.8-fold more in short- than in long-term survivors (Figure 1).

Clusterin Expression Correlates with Poor Survival and Chemoresponse

To evaluate whether clusterin expression correlates with poor survival, clusterin immunostaining was performed on an independent set of 62 ovarian cancer samples from patients with stage III, grade 3 serous ovarian cancers. After adjustment for optimal debulking and age, Cox regression analyses showed that high clusterin expression correlates with poor survival [hazard ratio (HR), 1.07; 95% confidence interval (CI), 1.002–1.443; P = .04]. In addition, using median weight scores as cutoffs, Kaplan-Meier survival analyses showed that patients with positive clusterin expression had significantly reduced survival times in both optimally and suboptimally debulked cases (P = .001 and P = .0096, respectively; Figure 2A). Normal ovarian surface epithelial cells had no immunostaining of clusterin. The immunostaining intensities of clusterin (weight score) in the nonresponder group was significantly (P = .019) higher than those in the responder group (Figure 2B).

Figure 2.

Figure 2

Open in a new tab

Clusterin expression correlates with overall survival and chemoresponse. (A) Kaplan-Meier survival curves for patients with stage III, high-grade serous ovarian adenocarcinomas. Patients who had tumors with positive clusterin expression (+) had significantly poorer prognoses than those who had negative clusterin expression (-), whether their debulking was optimal (i.e., the largest residual tumor was <2 cm) or suboptimal (i.e., the largest residual tumor was >2 cm). (B) Box-plot showing clusterin protein expression in chemoresponders and nonresponders. The box is bounded above and below by the 75th and 25th percentiles, and the median is the line in the box. Whiskers are drawn to the nearest value not beyond a standard span from the quartiles; points beyond (outliers) are drawn individually, where the standard span is 1.5 x (interquartile range).

Clusterin Expression Correlates with Paclitaxel Resistance in Ovarian Cancer Cell Lines

To test whether high clusterin expression is associated with chemoresistance, we determined the level of clusterin protein expression and paclitaxel IC50 values in 13 ovarian cancer cell lines. The Western blot showed that four cell lines (PEO4, RMUG-S, OVCA822, and TOV21G) had very high levels of clusterin expression (Figure 3A) and that these cell lines were the most resistant to paclitaxel. A significant correlation between clusterin expression and paclitaxel IC50 is illustrated in Figure 3B (Spearman's test, R2 = 0.55, P = .004). We have also determined the IC50 values for cisplatin, carboplatin, topotecan, and doxorubicin for these cell lines; however, no significant correlation was found.

Figure 3.

Figure 3

Open in a new tab

Correlation of clusterin expression and paclitaxel resistance. (A) Western blot analysis of clusterin expression in 13 epithelial ovarian cancer cell lines. (B) Scatter plot showing correlation between clusterin expression and paclitaxel IC50 for the 13 ovarian cancer cell lines tested. High IC50 values denote paclitaxel resistance, whereas low IC50 values indicate paclitaxel sensitivity. A Spearman correlation test showed that clusterin overexpression significantly correlated with IC50 for paclitaxel (R2 = 0.55, P = .004).

Up-regulation of Clusterin Expression Increases Paclitaxel Resistance of SKOV3 Cells

Paclitaxel-sensitive SKOV3 cells, which express low levels of clusterin, were stably transfected with a full-length human clusterin cDNA construct or with the vector alone. Cells transfected with the stable SKOV3/pcDNA3.1/CLU expressed high levels of clusterin and were more resistant to paclitaxel than those transfected with the vector control SKOV3/pcDNA3.1 (Figure 4, A and B; P < .05).

Figure 4.

Figure 4

Open in a new tab

Increased paclitaxel resistance in clusterin transfectants. (A) Western blot analysis showing overexpression of clusterin in SKOV3 cells stably transfected with plasmid pcDNA3.1 with clusterin (cDNA3.1-Clu). (B) SKOV3-cDNA3.1-Clu stably transfected expressing clusterin had higher survival rates than the control cells SKOV3-vector, which were transfected with pcDNA3.1 vector only.

Exogenous Clusterin Enhances Paclitaxel Resistance of SKOV3 Ovarian Cancer Cells

Because clusterin is a secreted protein, we evaluated the effects of exogenous clusterin on paclitaxel resistance. In the presence of paclitaxel, the number of surviving SKOV3 cells increased with the amount of exogenous clusterin (Figure 5; P < .05). As a control, we also tested the effect of clusterin alone on cell proliferation. There is no significant effect on cell proliferation when SKOV3 was treated with 7.5 µg/ml clusterin for 48 hours (Figure W1).

Figure 5.

Figure 5

Open in a new tab

Exogenous clusterin enhances the paclitaxel resistance of SKOV3 cells. Cell proliferation was measured by XTT assay in SKOV3 cells treated with 10-7 M paclitaxel (T) for 48 hours in the presence of the indicated concentrations of clusterin. The baseline used for calculating relative cell proliferation rate was set by adjusting the proliferation rate of paclitaxel-treated SKOV3 cells without exogenous clusterin to 1.

Clusterin Expression Knockdown by siRNA Reduces Paclitaxel Resistance of PEO4 Ovarian Cancer Cells

PEO4 is a paclitaxel-resistant cell line that expresses high levels of clusterin. Clusterin protein levels were significantly lower both 24 and 48 hours after PEO4 cells were transfected with clusterin siRNA (Figure 6A). No change in clusterin expression was observed in cells transfected with the luciferase siRNA control. After 24 hours of transfection with clusterin or luciferase siRNA, all PEO4 cells were treated with 10-7 M paclitaxel for 48 hours. A correlation between reduced clusterin expression and increased paclitaxel sensitivity was observed (Figure 6B). Moreover, paclitaxel-induced apoptosis also significantly increased when clusterin expression was knocked down in PEO4 cells (Figure W2). Similar results were obtained in two other paclitaxel-resistant cell lines, RMUG-S and TOV21G, which express high levels of clusterin (Figure W3).

Figure 6.

Figure 6

Open in a new tab

PEO4 became sensitive to paclitaxel after silencing clusterin expression with siRNA. (A) Western blot analyses show the time course of clusterin silencing in PEO4 cells transfected with siRNA for clusterin (+) or luciferase (-). After transfection, the cells were cultured for 2 days in medium containing 10-7 M paclitaxel. (B) Cell survival after paclitaxel treatment as determined by XTT assay.

Clusterin Expression Knockdown in Ovarian Cancer Cells by shRNA Reduces Tumor Size in Xenograft Mouse Models When Treated with Paclitaxel

PEO4 cells transduced with Lentivirus carrying clusterin or LacZ shRNA were xenografted into mice, and the effects of paclitaxel were determined. After 22 days, the tumors that had developed from the xenografted cells transduced with clusterin shRNA were significantly smaller than those of the control groups (Figure 7A). Decreased clusterin expression in xenografted tumor masses after autopsy was confirmed by Western blot analysis and densitometer quantification. The difference between shRNA-clusterin-treated tumor and shRNA-control was significant (twofold decrease) with P = .037 (Figure 7B).

Figure 7.

Figure 7

Open in a new tab

Effects of decreased clusterin expression on paclitaxel resistance in xenografted tumors. (A) When xenografts in nude mice resulting from the inoculation of paclitaxel-resistant cells transduced with shRNA for clusterin or the Lacz gene grew to approximately 100 mm3, mice were injected with 0.2 mg/0.1 ml per 10 g of paclitaxel intraperitoneally. Tumor volumes were measured twice every week. (B) Western blot analysis of clusterin expression from xenografted tumors from mouse autopsies. Tumors from three separate mice from both the shRNA-Lacz and shRNA-clusterin groups were analyzed.

Effect of Clusterin on Microtubule Binding

To evaluate whether clusterin confers paclitaxel resistance by preventing the binding of paclitaxel to microtubules, both paclitaxel-resistant PEO4 cells and paclitaxel-sensitive SKOV3 cells were treated with identical concentrations of paclitaxel-OG. PEO4 cells had minimal microtubule staining in the absence or presence of exogenous clusterin (Figure 8, A and B). In contrast, SKOV3 cells had strong cytoplasmic staining in the absence of clusterin, whereas no staining was observed in the presence of exogenous clusterin, suggesting that exogenous clusterin prevents the binding of paclitaxel to microtubules (Figure 8, C and D). To clarify the role of endogenous clusterin, we infected PEO4 cells with Lentivirus carrying shRNAclusterin (Lenti-shRNA-clusterin) to knockdown the endogenous clusterin expression. The result showed that PEO4 cells infected with Lenti-shRNA-clusterin had strong cytoplasmic stain by paclitaxel-OG, whereas PEO4 cells infected with Lentivirus vector control had undetectable staining similar to Figure 8, C and B, respectively (Figure W4).

Figure 8.

Figure 8

Open in a new tab

Exogenous clusterin prevents the binding of paclitaxel to microtubules in SKOV3 cells. PEO4 (A and B) and SKOV3 (C and D) cells cultured in chamber coverglasses were treated with 10-6 M OG-labeled paclitaxel in the absence (A and C) or presence (B and D) of 7.5 µg/ml clusterin for 30 minutes. Cells were also simultaneously incubated with the Hoechst stain for nuclei visualization, and fluorescent images were captured by an inverted fluorescence microscope equipped with a digital camera using the same setting. The binding of OG-labeled paclitaxel to microtubules is shown in panel C.

Binding Interactions between Clusterin and Paclitaxel

To evaluate whether clusterin binds specifically to paclitaxel, equimolar mixtures of clusterin or GST as a control and paclitaxel-OG, phalloidin-OG, or OG were incubated together and then fractionated by size exclusion chromatography. After incubation with clusterin, the elution profile of paclitaxel-OG adopted a pattern similar to that of clusterin alone (Figure 9, A and B). In contrast, after incubation with GST, the elution profile of paclitaxel-OG was similar to that of paclitaxel-OG alone, implying that under these conditions, the small molecule remained unbound. Similar results were obtained with two other control proteins, superoxide dismutase and hemoglobin (data not shown), and OG and phalloidin-OG, the control dyes (Figure W5).

Figure 9.

Figure 9

Open in a new tab

Coelution of clusterin and OG-labeled paclitaxel (paclitaxel- OG) in size exclusion chromatography. Solutions containing clusterin and paclitaxel-OG mixture, paclitaxel-OG alone, or a GST and paclitaxel-OG mixture (all at 5 mM) were fractionated by size exclusion chromatography in PEM buffer, and the OG fluorescence and A280 of the eluate were measured. The upper panel shows OG fluorescence as a function of elution time; the lower panel shows the corresponding A280 profile. The result shown is representative of three independent experiments.

Discussion

By comparing the expression profiles generated from short- and long-term ovarian cancer survivors, we found that clusterin expression was significantly higher in short- than in long-term survivors (Figure 1). Also, Kaplan-Meier survival analyses of results from clusterin immunostaining in 62 samples of late-stage, high-grade ovarian serous carcinomas indicated that the overexpression of clusterin correlates with poor survival (Figure 2A). This is consistent with reports that clusterin is a marker of poor prognosis in other cancer types [23–27]. Moreover, overexpression of clusterin also correlates with chemoresistance (Figure 2B).

Clusterin, also known as testosterone-repressed prostate message-2 (TRPM-2), sulfated glycoprotein-2 (SGP-2), SP-40, and apolipoprotein J (ApoJ), is a secreted heterodimeric glycoprotein encoded by a single gene on chromosome 8p21-p12. When analyzed by SDS-PAGE, clusterin migrates as a broad band at 70 to 80 kDa but has an actual mass of approximately 61 kDa (determined by mass spectrometry) [28]. Clusterin is present in various biologic fluids [29] and has been implicated in several physiological processes, including cell adhesion and aggregation [30], complement inhibition [31], lipid transport, membrane protection, sperm maturation, and endocrine secretion [31,32]. Clusterin expression has also been implicated in chemoresistance in several other cancer types [14,15]. Because the resistance of tumor cells to various available chemotherapeutic agents such as paclitaxel has been one of the major factors leading to poor survival in ovarian cancer patients, we therefore hypothesized that clusterin expression confers chemoresistance to ovarian cancer cells.

In this study, we demonstrated that clusterin expression correlated with paclitaxel resistance both in vitro and in vivo. We found that high levels of clusterin expression in ovarian cancer cell lines correlated with paclitaxel resistance, although a few cell lines with low levels of clusterin expression were also resistant to paclitaxel (Figure 3). These cell lines might use mechanisms other than overexpression of clusterin to confer paclitaxel resistance such as the expression of metabolizing enzyme cytochrome P450 or cellular efflux ABC transporter [33]. To demonstrate the role of clusterin in paclitaxel resistance, we manipulated the endogenous level of clusterin expression in a paclitaxel-sensitive cell line (SKOV3, low level of endogenous clusterin expression) and a paclitaxel-resistant cell line (PEO4, high level of endogenous clusterin expression). We found that paclitaxel-sensitive SKOV3 cells becamemore resistant to paclitaxel when endogenous clusterin expression level was up-regulated (Figure 4). Conversely, paclitaxel-resistant PEO4 cells became more sensitive to paclitaxel and more apoptotic in vitro and in vivo when endogenous clusterin expression level was down-regulated. These results indicated that high levels of endogenous clusterin expression were involved in the paclitaxel resistance of ovarian cancer cells.

In addition to the endogenous clusterin, the exogenous clusterin added in our experiments enhanced the paclitaxel resistance of SKOV3 cells, as indicated by the cells' increased survival rate and the reduction in paclitaxel-induced apoptosis (Figure 5). The extent of both the increased survival and the reduced apoptosis correlated with the increase in concentration of exogenous clusterin. This suggests that exogenous clusterin binds to extracellular paclitaxel and subsequently prevents it from entering the cells and exerting toxicity. To test the hypothesis that the exogenous clusterin secreted by ovarian cancer cells inhibits the binding of paclitaxel to microtubules inside the cells, we determined the level of microtubule-associated fluorescence in paclitaxel-OG-treated ovarian cancer cells with or without clusterin. Significant microtubule-associated fluorescence was observed in SKOV3 cells (Figure 8C) but not in PEO4 cells, which expressed high levels of endogenous clusterin (Figure 8, A and B). However, when SKOV3 cells were coincubated with both exogenous clusterin and paclitaxel-OG, negligible microtubule-associated fluorescence was observed (Figure 8D).

Using size exclusion chromatography to directly test whether clusterin bound specifically to paclitaxel in solution, we found that clusterin did bind to paclitaxel-OG but not to phalloidin-OG or OG alone and that GST and other control proteins did not bind to paclitaxel-OG. In addition, ELISA confirmed that clusterin bound with high affinity to paclitaxel but not at all to other therapeutic agents, such as cisplatin and carboplatin (data not shown). This is the first time we are aware of that the binding of clusterin and paclitaxel has been clearly demonstrated, although clusterin's role in chemoresistance has been suggested by previous reports in other cancer types [14–16]. Thus, our results suggest that clusterin, which is generally a secreted molecule, may protect cancer cells from paclitaxel toxicity by binding to the agent in the extracellular microenvironment and preventing it from entering the cell to exert its toxic effects.

In conclusion, clusterin overexpression significantly correlates with decreased survival in ovarian cancer patients, and clusterin confers resistance to paclitaxel-induced apoptosis in ovarian cancer cells. The mechanism of resistance may involve clusterin (potentially in both extra- and intracellular locations) complexing with paclitaxel to inactivate the agent's cytotoxic activity. Therefore, targeting the reduction of clusterin expression may be a worthwhile new therapeutic modality for treating ovarian tumors expressing high levels of clusterin.

Supplementary Material

Supplementary Figures and Tables
neo1009_0964SD1.pdf (274.3KB, pdf)

Abbreviations

shRNA

short hairpin RNA

CLU

clusterin

siRNA

small interfering RNA

Footnotes

1

This study was supported in part by the Dana Farber/Harvard Ovarian Specialized Programs of Research Excellence (SPORE) P50CA105009, the M.D. Anderson Cancer Center SPORE P50CA83639, and R33CA103595 from the National Institutes of Health; the Department of Health and Human Services; the Gillette Center For Women's Cancer; the Adler Foundation, Inc.; the Edgar Astrove Fund; the Ovarian Cancer Research Fund, Inc.; the CatholicMedical Center Research Foundation (from program year 2006); and the 2006 Research Fund from St. Vincent's Hospital.

2

This article refers to supplementary materials, which are designated by Table W1 and Figures W1 to W5 and are available online at www.neoplasia.com.

References

  • 1.Boring CC, Squires TS, Tong T, Montgomery S. Cancer statistics, 1994. CA Cancer J Clin. 1994;44:7–26. doi: 10.3322/canjclin.44.1.7. [DOI] [PubMed] [Google Scholar]
  • 2.Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics, 1998. CA Cancer J Clin. 1998;48:6–29. doi: 10.3322/canjclin.48.1.6. [DOI] [PubMed] [Google Scholar]
  • 3.Omura GA, Brady MF, Homesley HD, Yordan E, Major FJ, Buchsbaum HJ, Park RC. Long-term follow-up and prognostic factor analysis in advanced ovarian carcinoma: the Gynecologic Oncology Group experience. J Clin Oncol. 1991;9:1138–1150. doi: 10.1200/JCO.1991.9.7.1138. [DOI] [PubMed] [Google Scholar]
  • 4.Barlund M, Forozan F, Kononen J, Bubendorf L, Chen Y, Bittner ML, Torhorst J, Haas P, Bucher C, Sauter G, et al. Detecting activation of ribosomal protein S6 kinase by complementary DNA and tissue microarray analysis. J Natl Cancer Inst. 2000;92:1252–1259. doi: 10.1093/jnci/92.15.1252. [DOI] [PubMed] [Google Scholar]
  • 5.Wang T, Hopkins D, Schmidt C, Silva S, Houghton R, Takita H, Repasky E, Reed SG. Identification of genes differentially over-expressed in lung squamous cell carcinoma using combination of cDNA subtraction and microarray analysis. Oncogene. 2000;19:1519–1528. doi: 10.1038/sj.onc.1203457. [DOI] [PubMed] [Google Scholar]
  • 6.Xu J, Stolk JA, Zhang X, Silva SJ, Houghton RL, Matsumura M, Vedvick TS, Leslie KB, Badaro R, Reed SG. Identification of differentially expressed genes in human prostate cancer using subtraction and microarray. Cancer Res. 2000;60:1677–1682. [PubMed] [Google Scholar]
  • 7.DeRisi J, Penland L, Brown PO, Bittner ML, Meltzer PS, Ray M, Chen Y, Su YA, Trent JM. Use of a cDNA microarray to analyse gene expression patterns in human cancer [see comments] Nat Genet. 1996;14:457–460. doi: 10.1038/ng1296-457. [DOI] [PubMed] [Google Scholar]
  • 8.DeRisi JL, Iyer VR, Brown PO. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science. 1997;278:680–686. doi: 10.1126/science.278.5338.680. [DOI] [PubMed] [Google Scholar]
  • 9.Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science. 1999;286:531–537. doi: 10.1126/science.286.5439.531. [DOI] [PubMed] [Google Scholar]
  • 10.Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000;403:503–511. doi: 10.1038/35000501. [DOI] [PubMed] [Google Scholar]
  • 11.Miyake H, Gleave M, Kamidono S, Hara I. Overexpression of clusterin in transitional cell carcinoma of the bladder is related to disease progression and recurrence. Urology. 2002;59:150–154. doi: 10.1016/s0090-4295(01)01484-4. [DOI] [PubMed] [Google Scholar]
  • 12.Redondo M, Villar E, Torres-Munoz J, Tellez T, Morell M, Petito CK. Overexpression of clusterin in human breast carcinoma. Am J Pathol. 2000;157:393–399. doi: 10.1016/S0002-9440(10)64552-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Sensibar JA, Sutkowski DM, Raffo A, Buttyan R, Griswold MD, Sylvester SR, Kozlowski JM, Lee C. Prevention of cell death induced by tumor necrosis factor alpha in LNCaP cells by overexpression of sulfated glycoprotein-2 (clusterin) Cancer Res. 1995;55:2431–2437. [PubMed] [Google Scholar]
  • 14.Gleave ME, Miyake H, Zellweger T, Chi K, July L, Nelson C, Rennie P. Use of antisense oligonucleotides targeting the antiapoptotic gene, clusterin/testosterone-repressed prostate message 2, to enhance androgen sensitivity and chemosensitivity in prostate cancer. Urology. 2001;58:39–49. doi: 10.1016/s0090-4295(01)01241-9. [DOI] [PubMed] [Google Scholar]
  • 15.Miyake H, Hara I, Kamidono S, Gleave ME. Synergistic chemsensitization and inhibition of tumor growth and metastasis by the antisense oligodeoxynucleotide targeting clusterin gene in a human bladder cancer model. Clin Cancer Res. 2001;7:4245–4252. [PubMed] [Google Scholar]
  • 16.Miyake H, Nelson C, Rennie PS, Gleave ME. Acquisition of chemoresistant phenotype by overexpression of the antiapoptotic gene testosterone-repressed prostate message-2 in prostate cancer xenograft models. Cancer Res. 2000;60:2547–2554. [PubMed] [Google Scholar]
  • 17.Bonome T, Lee JY, Park DC, Radonovich M, Pise-Masison C, Brady J, Gardner GJ, Hao K, Wong WH, Barrett JC, et al. Expression profiling of serous low malignant potential, low-grade, and high-grade tumors of the ovary. Cancer Res. 2005;65:10602–10612. doi: 10.1158/0008-5472.CAN-05-2240. [DOI] [PubMed] [Google Scholar]
  • 18.Li C, Wong WH. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci USA. 2001;98:31–36. doi: 10.1073/pnas.011404098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001;98:5116–5121. doi: 10.1073/pnas.091062498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kim JH, Skates SJ, Uede T, Wong KK, Schorge JO, Feltmate CM, Berkowitz RS, Cramer DW, Mok SC. Osteopontin as a potential diagnostic biomarker for ovarian cancer. JAMA. 2002;287:1671–1679. doi: 10.1001/jama.287.13.1671. [DOI] [PubMed] [Google Scholar]
  • 21.Mok SC, Chao J, Skates S, Wong KK, Yiu GY, Muto MG, Berkowitz RS, Cramer DW. Prostasin, a potential serum marker for ovarian cancer, identified through microarray technology. J Natl Cancer Inst. 2001;93:1458–1464. doi: 10.1093/jnci/93.19.1458. [DOI] [PubMed] [Google Scholar]
  • 22.Wilson MR, Easterbrook-Smith SB. Clusterin is a secreted mammalian chaperone. Trends Biochem Sci. 2000;25:95–98. doi: 10.1016/s0968-0004(99)01534-0. [DOI] [PubMed] [Google Scholar]
  • 23.Miyake H, Gleave M, Kamidono S, Hara I. Overexpression of clusterin in transitional cell carcinoma of the bladder is related to disease progression and recurrence. Urology. 2002;59:150–154. doi: 10.1016/s0090-4295(01)01484-4. [DOI] [PubMed] [Google Scholar]
  • 24.Kruger S, Ola V, Fischer D, Feller AC, Friedrich M. Prognostic significance of clusterin immunoreactivity in breast cancer. Neoplasma. 2007;54:46–50. [PubMed] [Google Scholar]
  • 25.Miyake H, Hara S, Arakawa S, Kamidono S, Hara I. Over expression of clusterin is an independent prognostic factor for nonpapillary renal cell carcinoma. J Urol. 2002;167:703–706. doi: 10.1016/S0022-5347(01)69130-4. [DOI] [PubMed] [Google Scholar]
  • 26.Miyake H, Yamanaka K, Muramaki M, Kurahashi T, Gleave M, Hara I. Enhanced expression of the secreted form of clusterin following neo-adjuvant hormonal therapy as a prognostic predictor in patients undergoing radical prostatectomy for prostate cancer. Oncol Rep. 2005;14:1371–1375. [PubMed] [Google Scholar]
  • 27.Zhang S, Zhang D, Zhu Y, Guo H, Zhao X, Sun B. Clusterin expression and univariate analysis of overall survival in human breast cancer. Technol Cancer Res Treat. 2006;5:573–578. doi: 10.1177/153303460600500604. [DOI] [PubMed] [Google Scholar]
  • 28.Kapron JT, Hilliard GM, Lakins JN, Tenniswood MP, West KA, Carr SA, Crabb JW. Identification and characterization of glycosylation sites in human serum clusterin. Protein Sci. 1997;6:2120–2133. doi: 10.1002/pro.5560061007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Aronow BJ, Lund SD, Brown TL, Harmony JA, Witte DP. Apolipoprotein J expression at fluid-tissue interfaces: potential role in barrier cytoprotection. Proc Natl Acad Sci USA. 1993;90:725–729. doi: 10.1073/pnas.90.2.725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Silkensen JR, Skubitz KM, Skubitz AP, Chmielewski DH, Manivel JC, Dvergsten JA, Rosenberg ME. Clusterin promotes the aggregation and adhesion of renal porcine epithelial cells. J Clin Invest. 1995;96:2646–2653. doi: 10.1172/JCI118330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Rodenburg KW, Kjoller L, Petersen HH, Andreasen PA. Binding of urokinase-type plasminogen activator-plasminogen activator inhibitor-1 complex to the endocytosis receptors alpha2-macroglobulin receptor/low-density lipoprotein receptor-related protein and very-low-density lipoprotein receptor involves basic residues in the inhibitor. Biochem J. 1998;329:55–63. doi: 10.1042/bj3290055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Jenne DE, Tschopp J. Clusterin: the intriguing guises of a widely expressed glycoprotein. Trends Biochem Sci. 1992;17:154–159. doi: 10.1016/0968-0004(92)90325-4. [DOI] [PubMed] [Google Scholar]
  • 33.DeLoia JA, Zamboni WC, Jones JM, Strychor S, Kelley JL, Gallion HH. Expression and activity of taxane-metabolizing enzymes in ovarian tumors. Gynecol Oncol. 2008;108:355–360. doi: 10.1016/j.ygyno.2007.10.029. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figures and Tables
neo1009_0964SD1.pdf (274.3KB, pdf)

Articles from Neoplasia (New York, N.Y.) are provided here courtesy of Neoplasia Press

RESOURCES