Neutrophil extracellular traps (NETs) reduce the diffusion of doxorubicin which may attenuate its ability to induce apoptosis of ovarian cancer cells

Abstract

Purpose: Although neutrophil extracellular traps (NETs) are present in various tumors, their roles in tumor biology have not been clarified yet. In this study, we examined how NETs affect the pharmacokinetics and effects of doxorubicin (DOX).

Methods: NETs were generated by neutrophils stimulated with phorbol 12-myristate 13-acetate (PMA) or lipopolysaccharide (LPS). DOX was added to NETs and their distribution was observed under fluorescein microscopy, and the diffusion of DOX through 3 μM pores from lower to upper chambers was evaluated with a fluorescence-based assay. Ovarian cancer cells, KOC-2S and SKOV3, were embedded in collagen gel droplets and cultured in 3D way and their apoptosis was examined with flow cytometry.

Results: DOX was mostly co-localized with NETs. The transfer of DOX to upper chambers increased over time, which was significantly decreased by the presence of neutrophils stimulated with PMA or LPS in the lower chamber. DOX outside of the gel increased the rates of annexin V (+) apoptotic cells, which were significantly reduced by the addition of LPS-stimulated neutrophils in media both in KOC-2S and SKOV3. The reduced diffusion and apoptosis were mostly restored by the destruction of the NETs structure with 1000 u/ml DNAse I.

Conclusion: NETs efficiently trap and inhibit the diffusion of DOX which may attenuate its ability to induce apoptosis of ovarian cancer cells. Degradation of NETs with DNAse I may augment the response of ovarian cancer to DOX.

Keywords: Chemosensitivity; Doxorubicin; Neutrophil extracellular traps; Pharmacokinetics.

Figure 1

Fluorescein images of doxorubicin (DOX) and neutrophil extracellular traps (NETs). NETs were generated by stimulating neutrophils with 2 μM PMA (A–D) or 10 μg/ml LPS (E–H) and cultured in poly-L-lysine-coated plates as described in Materials and Methods. Then, DOX was added at a final concentration of 20 μM and incubated with NETs for 30 min. After gentle washing with warmed media, SYTOX green (50 nM) was added. As controls, DOX was added to freshly isolated neutrophils without washing (I–L). Images of bright fields (A, E, I). DOX was visualized using fluorescence microscopy with the optical wavelength filter for Tetramethyl rhodamine (TRICI) (B, F, J). NETs structures in the same field were observed with a wavelength filter for FITC (C, G, K) and the 2 fluorescein images were superimposed (D, H, L). Bars show a length of 100 μM.

Figure 2

Diffusion of doxorubicin (DOX) through culture inserts with 3 μM pores. Neutrophils (5 × 10⁶) stimulated with 2 μM PMA or 10 μg/ml LPS were suspended in 2 mL of HBSS without phenol red and placed in the bottom chambers and cultured for 4 h to form NETs. Freshly isolated neutrophils (5 × 10⁶/2 mL) were placed at 4 °C and used as unstimulated neutrophils. DOX (10 μM) was added and incubated for another 30 min. Then, culture inserts with pores containing 2 mL of HBSS were placed in the bottom chambers. After 1 (A) and 3 (B) hours, 100 μl of medium was collected from the upper chambers and fluorescence intensity measured. The relative ratios of the fluorescein intensities were calculated compared to the control well which did not contain neutrophils. Data are shown as mean ± standard deviation in 3 different experiments. ∗: p < 0.05, ∗∗: p < 0.01.

Figure 3

The effects of NETs on the diffusion of doxorubicin (DOX). Neutrophils (1 × 10⁷/4 ml) stimulated with PMA (A) or LPS (B) were placed in the bottom chamber as described in Figure 2 legend and DOX added at a final concentration of 10 μM and incubated for another 30 min. In some wells, DNAse I was added at a final concentration of 1000 u/ml at 30 min before the addition of DOX. Then, culture inserts containing 2 mL of HBSS were placed in the bottom chambers and auto fluorescence intensities of DOX in the upper chamber measured at the indicated time points. In each set of experiments, the relative ratios of fluorescein intensities were calculated against the value of the samples measured at 1 h after incubation in control wells which did not contain NETs and DNAse I. Data are shown as mean ± standard deviation in 3 (PMA) and 3 (LPS) different experiments. ∗: p < 0.05, ∗∗: p < 0.01.

Figure 4

Peritoneal tumors of SKOV-3 were induced as described in Materials and Methods, and similar sized tumors (approximately 3∼5 mm in diameters) were soaked in 50 mM DOX diluted in 4 ml of HBSS buffer with unstimulated (A) or PMA-stimulated (B) neutrophils in 15 ml tube. In (C), DNase I (1000 u/ml) was added with PMA-stimulated neutrophils at the start of the experiment. After 3 h, the tumors were taken out, fixed with dry-iced acetone and 10-μM cryostat sections of post-fixed frozen samples were created. After the counterstaining the nuclei with DAPI, the infiltration of DOX from the tumor surface was evaluated with the detection of autofluorescence under fluorescence microscopy (BZ8000; Keyence, Osaka, Japan). Figures show the merged images for DOX (red) and DAPI (Blue).

Figure 5

KOC-2S or SKOV3 cells were embedded in collagen gel droplets as described in Materials and Methods and cultured in 2 ml media containing 15 μM doxorubicin (DOX) with neutrophils (1 × 10⁷) and/or DNAse I (1000U/ml). NETs (-) and (+) show the data in the presence of unstimulated and LPS-stimulated neutrophils, respectively. After 12 h, apoptotic cells were examined with FACSCalibur. (A) Representative FACS Profiles of KOC-2S (B) Data are shown as mean ± standard deviation in triplicate from one of 4 (KOC-2S) and 3 (SCOV-3) different experiments. ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001.