Mucin impedes cytotoxic effect of 5-FU against growth of human pancreatic cancer cells: overcoming cellular barriers for therapeutic gain

Abstract

Mucins are high molecular weight glycoproteins expressed on the apical surface of normal epithelial cells. In cancer disease mucins are overexpressed on the entire cellular surface. Overexpression of MUC1 mucin in pancreatic tumours has been correlated with poor patient survival. Current chemotherapeutic approaches such as 5-fluorouracil (5-FU) has produced limited clinical success. In this study we investigated the role of mucin in cytotoxic drug treatment to determine whether the extracellular domain of mucin impedes cytotoxic drug action of 5-FU. Human pancreatic cancer cells revealed high and relatively moderate MUC1 levels for Capan-1 and HPAF-II, respectively, compared to MUC1 negative control (U-87 MG glioblastoma) that showed relatively non-specific anti-MUC1 uptake. Benzyl-alpha-GalNAc (O-glycosylation inhibitor) was used to reduce mucin on cell surfaces, and neuraminidase was used to hydrolyse sialic acid at the distal end of carbohydrate chains. Benzyl-alpha-GalNAc had no effect on cell morphology or proliferation at the concentrations employed. The inhibition of O-glycosylation resulted in significant 5-FU antiproliferative activity against Capan-1 and HPAF-II, but not against U-87 MG. However, the exposure of cells to neuraminidase failed to improve the cytotoxic action of 5-FU. Our experimental findings suggest that the overexpression of mucin produced by human pancreatic tumours might limit the effectiveness of chemotherapy.

Figure 1

Immunofluorescence staining of cells using anti-MUC1 antibody (CD227). (A, B) DIC microscopy images show clusters of Capan-1 and HPAF-II cell lines and (C) DIC images show U-87 MG (negative control) cells without clustering. (D, E, F) Fluorescence microscopy images show relative extent of FITC-conjugated anti-MUC1 antibody (CD227) associated with cells. (G, H, I) Superimposed images confirm areas of antibody location with respect to each cellular cluster (× 20 magnification).

Figure 2

Determining the maximum non-toxic concentration of benzyl-α-GalNAc. (A) Percent cell viability of Capan-1 (○), HPAF-II (▪) and U-87 MG (Δ) cells was determined after exposing 1 × 10⁴ cells per ml to different concentrations of benzyl-α-GalNAc for 72 h. Percent viability of Capan-1 cells was significantly lower at concentrations greater than 0.4 mg ml⁻¹ (^#P<0.05 vs 0.8 mg ml⁻¹); whereas the viability of HPAF-II cells was significantly lower beyond 0.8 mg ml⁻¹ concentration (^*P<0.05 vs 2.0 mg ml⁻¹); no cell death was observed for U-87 MG cells upto 2.0 mg ml⁻¹ concentration. (B) Real-time RT–PCR showed MUC1 mRNA expression levels for Capan-1 >HPAF-II >U-87 MG. The exposure of these cells to benzyl-α-GalNAc (+) did not induce alteration in MUC1 mRNA expression levels when compared to cells not exposed to benzyl-α-GalNAc (−). (C) DIC microscopy images show morphology of cells that were not exposed to benzyl-α-GalNAc (−) compared to structure of cells following 72 h exposure to benzyl-α-GalNAc (+) (0.4 mg ml⁻¹ for Capan-1; 0.8 mg ml⁻¹ for HPAF-II and U-87 MG) (× 20 magnification).

Figure 3

Inhibition of MUC1 mucin O-glycosylation. Approximately 2 × 10⁴ cells per ml of media were exposed to benzyl-α-GalNAc for 24, 48 and 72 h followed by 24 h incubation with FITC-conjugated anti-MUC1 (CD227) monoclonal antibody (4 μl well⁻¹) at 37°C. The relative fluorescence intensities correlate with affinity of cells pretreated with benzyl-α-GalNAc (+) for CD227 compared to control (benzyl-α-GalNAc (−)) cells. Fluorescence intensities for antibody associated with (A) Capan-1 and (B) HPAF-II cells exposed to benzyl-α-GalNAc was significantly higher for 24, 48 and 72 h exposure time (^*P⩽0.001) as compared to control (benzyl-α-GalNAc (−)). (C–D) Fluorescence-activated cell-sorting analysis for (C) Capan-1, (D) HPAF-II and (E) U-87 MG when exposed to benzyl-α-GalNAc (+) and compared to cells that were not previously exposed to benzyl-α-GalNAc. The specific peaks and lines to which the peaks are assigned are in the insert of the panel for FACS analysis.

Figure 4

Effect of MUC1 mucin O-glycosylation on the cytotoxic action of 5-FU. We exposed (∼1 × 10⁴) cells to benzyl-α-GalNAc for 48 h followed by 24 h of treatment with 5-FU. (A) Percent viability of control cells (exposed to benzyl-α-GalNAc alone) was approximately 100%, whereas viability of HPAF-II cells exposed to benzyl-α-GalNAc followed by 5-FU treatment decreased with increasing concentrations of benzyl-α-GalNAc, and was significant at concentrations ⩾0.4 mg ml⁻¹ (^*P⩽0.001 vs 5-FU treatment alone). (B) Percent viability of Capan-1 and HPAF-II cells were significantly lower (^*P⩽0.001) for cells exposed to benzyl-α-GalNAc followed by 5-FU treatment, compared to cells treated with 5-FU alone. The percent viability of U-87 MG cells (negative control) was similar for cells exposed to benzyl-α-GalNAc followed by 5-FU compared to 5-FU treatment alone.

Figure 5

Cell proliferation and morphology post-removal of benzyl-α-GalNAc. (A) Capan-1 and (B) HPAF-II cells (1 × 10⁴ cells per ml) were exposed to 0.4 and 0.8 mg ml⁻¹ of benzyl-α-GalNAc, respectively. Following 48 h exposure to benzyl-α-GalNAc the cells were washed with 1 × PBS and allowed to grow for next 24, 48 and 72 h in fresh media. The percent viability of cells was measured at each time point following exposure to benzyl-α-GalNAc (open bars, □) and compared with percent viability of cells not exposed to benzyl-α-GalNAc (closed bars, ▪). (C–D) Viability of Capan-1 and HPAF-II cells at 100% 24, 48 and 72 h post-removal of benzyl-α-GalNAc, and DIC microscopy of (C) Capan-1 and (D) HPAF-II cells post-removal of inhibitor (× 20 magnification).

Figure 6

Effect of sialic acid removal on cytotoxic activity of 5-FU. Approximately 1 × 10⁴ HPAF-II cells per ml of growth medium were exposed to neuraminidase for 1 h at 37°C. (A) Cell viability was 100% upto 0.05 U ml⁻¹ concentration of neuraminidase. (B) FACS analysis showing decrease in fluorescence intensity (correlating to association of FITC-conjugated MAA lectin to the sialic acid residues) for cells exposed to (+) neuraminidase compared to cells without exposure to (−) neuraminidase. (C) FACS analysis showing no change in fluorescence peaks for CD227 association with or without neuraminidase treatment. (D) When cells were previously exposed to neuraminidase (+), the effect of 5-FU on percent of cell viability was not significantly different from the control (neuraminidase (−)).