Objective
Clear cell carcinoma of the ovary is a distinct subtype of epithelial cancer associated with chemoresistance and poor outcome compared with serous papillary carcinomas. Resistance to paclitaxel has been linked to serous papillary overexpression of class III β-tubulin in several human cancers but inadequately characterized among clear cell carcinoma of the ovary. Chemoresistance has also been variably linked to the drug efflux pump p-glycoprotein. Epothilones are microtubule-stabilizing agents with putative activity in paclitaxel-resistant malignancies. In this study, we clarify the relationship between class III β-tubulin and p-glycoprotein expression in clear cell carcinoma of the ovary, clinical outcome, and in vitro responsiveness to patupilone and paclitaxel.
Study Design
Class III β-tubulin and p-glycoprotein were quantified by real time polymerase chain reaction in 61 fresh-frozen tissue samples and 11 cell lines. Expression by polymerase chain reaction was correlated with immunohistochemistry and overall survival. IC 50 was determined using viability/metabolic assays. Impact of class III β-tubulin down-regulation on IC 50 was assessed with small interfering RNAs.
Results
Clear cell carcinoma of the ovary overexpressed class III β-tubulin and p-glycoprotein relative to serous papillary carcinomas carcinomas in fresh-frozen tissues and cell lines. Class III β-tubulin immunohistochemistry reflected real time polymerase chain reaction results and overexpression stratified patients by overall survival. P-glycoprotein correlated with in vitro paclitaxel resistance, but not clinical outcome. Clear cell carcinoma of the ovary were exquisitely sensitive to patupilone in a manner that correlated with class III β-tubulin expression.
Conclusion
Class III β-tubulin overexpression in clear cell carcinoma of the ovary discriminates poor prognosis, serves as a marker for sensitivity to patupilone, and may contribute to paclitaxel resistance. Immunohistochemistry reliably identifies tumors with overexpression of class III β-tubulin, and accordingly a subset of individuals likely to respond to patupilone.
Ovarian cancer represents the leading cause of death from gynecologic malignancy in the United States and most developed countries. Ovarian clear cell carcinoma (OCCC) is a distinct histologic subtype of epithelial cancer associated with endometriosis, paraneoplastic hypercalcemia, and thromboembolism. OCCC constitutes 3.7 to 25% of epithelial ovarian cancers, with a higher incidence in Asian women.
Compared with ovarian serous papillary carcinoma (OSPC), OCCC carries a poorer prognosis characterized by chemoresistance and propensity for recurrence. The Gynecologic Oncology Group found clear cell histology to be an independent predictor of poorer progression-free and overall survival (OS) in an investigation of 1896 patients with advanced epithelial ovarian cancer. In a separate comparison of women with optimally debulked stage III disease, 52% with clear cell histology progressed on platinum-based chemotherapy compared with 29% of those with serous histology ; median survival was 12 vs 20 months, respectively. Paclitaxel in combination with platinum chemotherapy represents the current standard of care for adjuvant therapy after surgical staging for this aggressive disease. Initial response rates of OCCC to this regimen vary from 22-56%, which compare unfavorably with response rates that exceed 70% for OSPC.
Resistance to paclitaxel has been linked to overexpression of class III β-tubulin , 1 of 9 β-isotypes capable of heterodimerizing with α subunits to form microtubules critical to cell division. Paclitaxel binds preferentially to class I β-tubulin isotypes, which differ from class III β-tubulin at several taxol binding sites. Class III β-tubulin expression reduces the rate of microtubule assembly and correlates with poorer OS in non-small cell lung, breast, and unknown primary cancers. Although this clinicopathologic association has also been demonstrated in ovarian cancers, existing studies focus on serous histologies or use semiquantitative methods ; clear cell carcinomas have been undercharacterized.
Epothilones (EPOxide THIazoLe ketONEs) are novel microtubule-stabilizing macrolides isolated from Sorangium cellulosum with activity in paclitaxel-resistant malignancies putatively because of their unique ability to bind class III and I isotypes with at least equal affinity. Epothilones generally do not share with paclitaxel overlapping mechanisms of resistance, and are not substrates for the drug efflux pump p-glycoprotein encoded by the gene ABCB1 , the overexpression of which has been linked to multidrug resistance to taxanes, anthracyclines, vinca alkaloids, and epipodofilotoxins. Despite encouraging in vitro data and strong biologic plausibility, p-glycoprotein expression often does not correlate with clinical outcome and clinical trials of p-glycoprotein inhibitors have failed (reviewed in ). This underscores the need to identify novel biomarkers that predict clinical response.
In this study, we sought to (1) quantify class III β-tubulin and p-glycoprotein expression in a large number of OCCC in comparison to OSPC by quantitative real time polymerase chain reaction (qRT-PCR) within solid tissues and cell lines (2) describe the association of class III β-tubulin and p-glycoprotein with in vitro chemoresponsiveness to epothilone B (patupilone) and paclitaxel (3) characterize the contribution of class III β-tubulin to paclitaxel resistance (4) correlate class III β-tubulin immunohistochemistry with qRT-PCR expression level (5) examine the prognostic implications of class III β-tubulin overexpression on clinical outcome.
Materials and Methods
Tissue procurement and establishment of cell lines
As approved by institutional review boards at Yale University/University of Brescia, fresh frozen tissues were obtained at time of primary debulking. Tumors were staged according to criteria of the AJCC 7th ed./FIGO. Per WHO guidelines, those that contained <10% of a second component were considered “pure;” those which contained >10% were considered “mixed.” Cell lines were established as described previously. Tumor cells were maintained as a monolayer in RPMI-1640 supplemented with 10% fetal bovine serum (FBS; Gemini, Woodland, CA), 1% penicillin-streptomycin (Mediatech, Manassas, VA), and 0.3% amphotericin B (Invitrogen, Carlsbad, CA). Cells were incubated at 37°C in a humidified atmosphere of 95% air/5% CO 2.
RNA extraction, purification, and reverse-transcription
Total RNA extraction was performed using AllPrep DNA/RNA/Protein Minikit (Qiagen, Germantown, MD) per the manufacturer’s protocol. qRT-PCR was performed with a 7500 RealTime PCR System (Applied Biosystems, Foster City, CA). RNA samples were treated with Turbo DNase enzyme (Turbo DNA-free kit; Applied Biosystems). Total RNA (5 μg) was reverse-transcribed using Superscript III first-strand cDNA synthesis (Invitrogen). Five microliters of reverse-transcribed RNA was amplified using Taqman Universal PCR Master Mix (Applied Biosystems). Class III β-tubulin (TUBB3; Hs00964962_g1) and p-glycoprotein (ABCB1; Hs00184491_m1) primers were obtained (Applied Biosystems), with GAPDH as an internal control (Hs99999905_m1). Gene expression was analyzed using the comparative threshold method and normalized to a common calibrator.
Characterization of cellular growth rate
Cells at ≥80% confluence were harvested by trypinsinization using 0.05% trypsin-EDTA (Invitrogen, Carlsbad, CA), counted on a hemocytometer, and plated at a density of 100,000 cells/2 mL medium. Cells were incubated as above and trypsinized at 24, 48, 72, and 96 hours. Viable cells were assessed by trypan blue exclusion. Doubling time (DT) was calculated for exponential growth according to DT = (hours elapsed) (ln2)/[ln(number cells t2 /number cells t1 )].
Chemosensitivity and viability assays
Patupilone (Novartis Pharma, Basel, Switzerland) and paclitaxel (T7402; Sigma-Aldrich, St. Louis, MO) were dissolved in dimethyl sulfoxide (Sigma-Aldrich) and stored at −20°C protected from light. For metabolic assays, cells at log phase of growth were seeded in 96-well plates at optimum density, incubated, and exposed to drug after 24 hours. MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium] (Promega Corporation, Madison, WI) was added after 48-72 hours of additional incubation. Absorbance at λ = 490 nm was measured with a VersaMax microplate reader (Molecular Devices, Sunnyvale, CA). For viability assays, cells at log phase of growth were seeded at optimum density and exposed to drug at 24 hours. After 48-72 hours of additional incubation, well contents were harvested in entirety, centrifuged then stained with either trypan blue for manual counts or propidium iodide (2 μL of 500 μg/mL solution in PBS, 0.1% sodium azide, 2% FBS) for flow cytometric counts. Unless stated otherwise, all IC 50 values presented have been determined by flow cytometry.
Immunohistochemistry
Slides were reviewed by a gynecologic pathologist (N.B.). Representative formalin-fixed paraffin-embedded tissue blocks were cut at 4 μm and stained with anticlass III β-tubulin monoclonal antibody (TUJ1; Covance, Berkeley, CA) at 1:500 according to manufacturer’s instructions. Appropriate positive/negative controls were used. Staining intensity was assessed using a semiquantitative system: 0 = negative; 1+ = focal weak staining; 2+ = diffuse weak/focal moderate staining; 3+ = diffuse moderate/focal strong staining; and 4+ = diffuse strong immunoreactivity. For mixed tumors, only the clear cell component was scored.
Small interfering RNA
Class III β-tubulin-specific validated small-interfering RNA oligonucleotides (5′>3′ CCAUUCUGGUGGUGGACCUGGAtt) were purchased (Silencer Select; Life Technologies, Carlsbad, CA). Cells were seeded at a concentration of 200,000/well and transfected with small interfering RNA (siRNA) duplexes at 10 nM with 5 μL Lipofectamine RNAiMAX (Invitrogen) in OptiMem (Life Technologies) and RPMI with 10% FBS. Mock transfections with nonspecific siRNA duplexes were used as negative controls. Cells were exposed to drug 24 hours after plating. Cells remained in culture for a total of 72 hours, after, which they were collected for RNA extraction/qRT-PCR and/or viability assays.
Statistical analyses
Dose response curves were analyzed using GraphPad Prism v.5.0 (GraphPad Software, Inc, LaJolla, CA). IC 50 was defined as concentration of drug required to achieve viability of 50% untreated control. IC 50 was determined through interpolation of sigmoidal curves fit with a Hill slope of −1.0. Kaplan-Meier curves were analyzed with log-rank-Mantel-Cox tests with a consistent hazard ratio. OS was calculated from date of surgery to date of last follow-up/death. All Student’s t tests/Pearson correlations employed 2-tails and assumed equal variance. Nonparametric Mann-Whitney analyses were applied to immunohistochemistry scores. For all analyses, P values < .05 were considered statistically significant. Unless stated otherwise, mean ± standard error are reported.
Results
Ovarian clear cell carcinomas overexpress class III β-tubulin in fresh-frozen tissues and cell lines by qRT-PCR
Fresh-frozen tissue specimens representing 26 OCCC and 35 OSPC were analyzed. A total of 4 OCCC and 7 OSPC primary cell lines were established. Source patient characteristics are provided in the Tables 1 and 2 . OCCC overexpressed class III β-tubulin relative to OSPC in both fresh-frozen tissues ( Figure 1 , A ; 154.9 ± 24.42 vs 89.97 ± 20.60, P = .046) and cell lines ( Figure 1 , B; 1948 ± 261.6 vs 645.1 ± 157.9, P = .001). Class III β-tubulin was not expressed at high levels in normal human ovarian surface epithelial cell lines (133.3 ± 22.37, n = 3). In fresh-frozen tissues, there was no difference in expression between mixed/pure OCCC (130.7 ± 29.42 vs 163.8 ± 45.26, P = .56), or between early/advanced stage disease (157.9 ± 29.21 vs 154.4 ± 35.98, P = .88).
| Tissues | OCCC (n = 26) | % | OSPC (n = 35) | % |
|---|---|---|---|---|
| Stage | ||||
| I | 8 | 30.8 | 0 | 0.0 |
| II | 5 | 19.2 | 3 | 8.6 |
| III | 12 | 46.2 | 26 | 74.3 |
| IV | 1 | 3.8 | 6 | 17.1 |
| Histology | ||||
| Pure | 19 | 73.1 | 35 | 100.0 |
| mixed | 7 | 26.9 | 0 | 0.0 |
| Race | ||||
| White | 25 | 96.2 | 33 | 94.3 |
| Black | 0 | 0.0 | 2 | 5.7 |
| Other | 1 | 3.8 | 0 | 0.0 |
| Age, mean (y) [range] | 63 | [34–85] | 60 | [24–90] |
| Cell lines | Age | Stage | Race |
|---|---|---|---|
| OSPC-1 | 52 | IIIC | White |
| OSPC-2 | 52 | IV | White |
| OSPC-3 | 58 | IV | White |
| OSPC-6 | 60 | IIIC | White |
| OSPC-7 | 64 | IIIC | White |
| OSPC-8 | 57 | IC | White |
| OSPC-9 | 57 | IIIC | White |
| OCCC-1 | 50 | IC | White |
| OCCC-2 | 79 | IC | White |
| OCCC-3 | 42 | II | White |
| OCCC-5 | 32 | IC | White |
Ovarian clear cell carcinomas overexpress p-glycoprotein in fresh-frozen tissues but not cell lines by qRT-PCR
P-glycoprotein expression was also higher in fresh-frozen tissues among OCC compared with OSPC (858.4 ± 287.41 vs 96.90 ± 64.40, P = .03) (not shown); this did not reach significance in the 11 cell lines examined (84.22 ± 48.82 vs 266.5 ± 136.5, P = .36).
Immunohistochemistry staining for class III β-tubulin correlates with qRT-PCR expression level
Approximately 20% of specimens included in the qRT-PCR analysis were also examined with immunohistochemistry (IHC) (4 OCCC, 8 OSPC). OCCC demonstrated stronger staining (mean IHC score between 2+ and 3+) relative to OSPC (mean IHC score between 0 and 1+) ( P = .03). qRT-PCR expression among these specimens was 223.0 ± 30.42 vs 97.01 ± 28.60 ( P = .02). Because we have previously demonstrated that immunohistochemistry reflects qRT-PCR in other aggressive histologies, such as uterine serous carcinoma, and because statistical significance was already achieved comparing the first 12 tumor specimens included in both the qRT-PCR and IHC analysis, we did not perform immunohistochemistry in the remaining specimens. Normal ovary did not express class III β-tubulin by IHC; this reflects qRT-PCR results of normal ovary, endometrium, and myometrium (32.63 ± 6.51, n = 3). Representative slides are shown in Figure 2 .
Ovarian clear cell carcinomas are highly sensitive to patupilone relative to ovarian serous carcinomas
Next, we performed chemosensitivity profiling of 3 OCCC and 3 OSPC cell lines. These cell lines were chosen for study specifically on the basis of their similarities in growth rates, as variations may contribute significantly to responsiveness to chemotherapy. OCCC were found to be highly sensitive to patupilone. As determined by flow cytometric assays of cell viability, IC 50 values for patupilone among OCCC and OSPC cell lines were 0.06 ± 0.01 nM and 1.01 ± 0.13 nM, respectively ( P = .01) ( Figure 3 , A ). Corresponding IC 50 values for paclitaxel were 12.21 ± 1.17 and 95.64 ± 75.34 ( P = .13). Thus, although these OCCC lines were not significantly more sensitive to paclitaxel than OSPC (7.8-fold), they were exquisitely more sensitive than OSPC to patupilone (16.8-fold). DTs for OCCC and OSPC were 24.26 ± 1.70 vs 35.32 ± 8.49 ( P = .28; not shown). Flow cytometric findings were generally robust and consistent across both metabolic assays and manual counts by trypan blue exclusion. Results of these independent validation studies are summarized in Figure 3 , B and C.
Chemosensitivity of ovarian clear cell carcinomas to patupilone varies with class III β-tubulin overexpression; chemoresistance to paclitaxel correlates with p-glycoprotein
To further investigate potential mechanisms underlying the chemoresistance patterns observed in these particular cell lines, we also examined p-glycoprotein expression and found that qRT-PCR expression correlated with paclitaxel resistance ( Figure 4 , A ), although class III β-tubulin expression correlated with sensitivity to patupilone ( Figure 4 , B). Across these 3 OCCC and 3 OSPC cell lines, there was higher class III β-tubulin expression among OCCC with a trend toward statistical significance (1817 ± 319.9 vs 834.2 ± 289.6, P = .09). P-glycoprotein expression was not different (98.81 ± 65.88 vs 209.2 ± 103.8, P = .42) and did not correlate with patupilone IC 50 ( P = .134; not shown).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
