Cisplatin Fails To Stimulate Production Of Reactive Oxygen Species and Release Of Neutrophil Extracellular Traps By Human Neutrophils In Vitro

Cisplatin is a commonly used antineoplastic agent for treatment of a broad range of cancers. Cisplatin-based treatment has been associated with a significant risk of venous thromboembolism. The mechanisms through which cisplatin contributes to a prothrombotic state remain unclear. Neutrophil extracellular traps (NETs) consist of web-like DNA–histone core decorated with granule proteins and are released from activated neutrophils in a process dependent on reactive oxygen species (ROS), in particular hypochlorous acid (HOCl). Recently, NETs have been shown to play an important role in initiation and propagation of venous thrombus in a number of animal models of deep vein thrombosis. The aim of this study was to investigate whether NETs may provide a potential link between cisplatin and venous thromboembolism.

To assess the effect of cisplatin on release of NETs by ex vivo human neutrophils isolated by positive immunomagnetic selection we visualised NETs release by confocal fluorescent microscopy and performed fluorimetric quantification of cell-free DNA (CFDNA) using either SYTOX Green nucleic acid stain (10 µM) or an ultrasensitive fluorescent assay Picogreen Quant IT (Invitrogen). In contrast to stimulation with phorbol 12-myristate 13-acetate (PMA) (25 nM),which resulted in 22 ng/10⁴neutrophils of detectable CFDNA, neither of these two assays could detect any significant release of CFDNA by human neutrophils exposed to cisplatin (15 µM) for 2 or 4 hours above baseline similar with vehicle control. Furthermore, confocal fluorescent microscopy imaging of neutrophils stained with non-cell permeable DNA dye SYTOX Red (Invitrogen) demonstrated no difference in NET formation between control and cisplatin treated human neutrophils. Thus we could not demonstrate that NETS are produced in response to cisplatin treatment.

In view of consistent reports that NET formation is ROS dependent we decided to investigate whether cisplatin exposure leads to production of ROS by human neutrophils. Few published studies into the effects of cisplatin on the production of ROS by human neutrophils in vitro offer conflicting results.

We used flow cytometry and fluorescent probe hydroethidine (HE) for detection of intercellular superoxide anion radical in HL60 granulocytic cells in the presence of cisplatin (up to 50 µM). Differentiation down the granulocytic lineage after stimulation with ATRA was confirmed by light microscopy and by flow cytometry. Capacity of differentiated HL60 cells to generate NET formation after PMA stimulation was confirmed by fluorescence microscopy. Cisplatin failed to augment the spontaneous production of ROS by ATRA differentiated HL60 cells. The number of viable ethidium-high cells in cisplatin treated group did not differ from the vehicle control indicating no detectable production of ROS in response to cisplatin. In contrast, positive control treatment with PMA (25 nM) and menadione (40 µM) resulted in 4- and 20-fold increase in viable ethidium-high population respectively.

ROS generation by human neutrophils was measured by a colorimetric assay for chlorination of extracellular taurine to determine if exposure to cisplatin results in the production of HOCl by human neutrophils in vitro. Treatment of resting neutrophils with cisplatin (15 µM) for 30 min or 120 min was not associated with an increase in the spontaneous production of HOCl above the baseline. Furthermore, the PMA (25 nM)-activated generation of HOCl production was not increased by pre-treating neutrophils with cisplatin indicating that there was no potentiation of ROS by pre-treatment with cisplatin.

Our results suggest that cisplatin fails to induce release of NETs or HOCl from human neutrophils in vitro. These negative findings seem to be at odds with the well described pro-oxidative actions of cisplatin. One possible explanation centres on reported findings that the pro-oxidative effects of cisplatin are dependent on the mitochondrial generation of ROS whilst the mitochondria-generated ROS appear not to be instrumental to NET formation. Therefore, we postulate that cisplatin may not be able to induce NET formation by human neutrophils, which are known to contain few mitochondria, due to a sub-threshold ROS signal. Therefore it appears that cisplatin-associated increased risk of venous thrombosis is unlikely to be mediated through NETs.

No relevant conflicts of interest to declare.