Paclitaxel sensitivity in relation to ABCB1 expression, efflux and single nucleotide polymorphisms in ovarian cancer

Abstract

ABCB1 (adenosine triphosphate-binding cassette transporter B1) mediates cellular elimination of many chemotherapeutic agents including paclitaxel, which is commonly used to treat ovarian cancer. A significant association between common single nucleotide polymorphisms (SNPs) in ABCB1 and progression-free survival has been reported in patients with ovarian cancer. Variable paclitaxel clearance due to genotype specific differences in ABCB1 activity in cancer cells and/or normal tissues may underlie the association. Using cell-based models, we evaluated the correlations between ABCB1 expression, polymorphisms, transporter activity and paclitaxel sensitivity in ovarian cancer (n = 10) and lymphoblastoid (n = 19) cell lines. Close associations between ABCB1 expression, transporter function and paclitaxel sensitivity were found in lymphoblastoid cell lines, although we could not demonstrate an association with common SNPs. In ovarian cancer cell lines, ABCB1 expression was low and the association between expression and function was lost. These results suggest that ABCB1 related survival difference in ovarian cancer patients is more likely to be due to differential whole body paclitaxel clearance mediated by normal cells rather than a direct effect on cancer cells.

Figure 1. Relative ABCB1 expression in lymphoblastoid and ovarian cancer cell lines.

ABCB1 mRNA expression was measured by quantitative RT-PCR in (a) 19 lymphoblastoid cell lines and (b) 10 ovarian cancer cell lines. All results were expressed relative to the level expressed in OVCAR-3. The expression in SK-OV-3, A2780 and OVCAR-3 in (a) were measured in parallel to compare the levels of relative expression between lymphoblastoid cell lines and ovarian cancer cell lines. The results presented were the geometric mean of values normalized to the three control genes HPRT, GUSB and PGK1.

Figure 2. Cellular accumulation and efflux of bodipy-FL-paclitaxel, a fluorescent ABCB1 substrate in lymphoblastoid and ovarian cancer cell lines measured by flow cytometry.

(a) Median bodipy-FL-paclitaxel accumulated in 13 LCLs and six ovarian cancer cell lines, by independent-samples median test; (b) Decreasing bodipy-FL-paclitaxel fluorescence intensity over a 10 minutes efflux time, in one LCL as a representative example; (c) A fitted curve generated by Solve function of Excel as compared to the actual curve of percentage fluorescence retained, in one LCL as a representative example; (d) Median efflux constant K_E in LCLs and ovarian cancer cell lines, by independent-samples median test. All results are mean of at least two independent experiments.

Figure 3. Inhibition of proliferation by paclitaxel in lymphoblastoid and ovarian cancer cell lines.

Effect of paclitaxel on cell proliferation was measured in (a) 19 lymphoblastoid cell lines and (b) 10 ovarian cancer cell lines using the CellTitre AQueous One Solution Cell Proliferation Assay, measured after 72 hrs drug exposure, represented as a percentage of vehicle treated control. Results are mean with standard error of mean of three independent experiments performed in triplicate.

Figure 4. Relationship between growth inhibitory effects of paclitaxel, ABCB1 expression and cellular drug accumulation and efflux.

Paclitaxel growth inhibitory effect (IC₅₀) was determined in 13 lymphoblastoid cell lines (a), (c) and (e) and 6 ovarian cancer cell lines (b), (d) and (f) by CellTitre AQueous One Solution Cell Proliferation Assay after 72 hrs drug exposure. ABCB1 mRNA expression (a) and (b) was measured by RT-PCR and bodipy-FL-paclitaxel accumulation (c) and (d) and efflux (e) and (f) were measured by flow cytometry. The correlation between paclitaxel IC₅₀ and ABCB1 expression (a) and (b), paclitaxel IC₅₀ and drug accumulation (c) and (d) and the correlation between IC₅₀ and drug efflux constant K_E (e) and (f) were determined by simple linear regression analysis.

Figure 5. Association between growth inhibitory effects of paclitaxel and ABCB1 genotypes.

Paclitaxel growth inhibitory effects (IC₅₀) were determined in 19 lymphoblastoid cell lines (a), (c) and (e) and 10 ovarian cancer cell lines (b), (d) and (f), by CellTitre AQueous One Solution Cell Proliferation Assay after 72 hrs drug exposure. ABCB1 genotype was determined by iPLEX genotyping on Sequenom's MassARRAY platform. Associations between IC₅₀ and ABCB1 genotypes at C1236T (a) and (b), G2677T/A (c) and (d) and C3435T (e) and (f) were analysed by Jonckheere-Terpstra test.