br After stimulation by BCG for h the
After stimulation by BCG for 2 h, the contour of neutrophils changed. Specifically, the cell membranes were shown to be matted under an optical microscope (Fig. 1A). The formation of NETs was confirmed by staining and visualized by CLSM and SEM. Under SEM, BCG-stimulated neutrophils were shown to be firmly attached to the slide and aggregated. In addition, the cellular morphology was char-acterized by flattening of the plasma membrane with extracellular ex-trusion of fibrillar material, which was consistent with the cellular morphology following PMA stimulation (Fig. 1B). CLSM demonstrated
(caption on next page)
Fig. 3. NETs prevented the viability of BC cell lines.
A, representative images of tumor FG-4592 (T24) treated with NETs for 24 h showed a dose-dependent influence on morphology. B, inhibition ratios of T24 cells, which were determined by CCK-8, increased with time and dose of NETs. C, at 48 h of incubation, the extent of growth did not decrease after boiled-NETs treatment, but significantly decreased after DNase-digested NETs treatment, with NETs one as the control. D, NETs exerted a cytotoxic eﬀect on tumor cells, and the eﬀect was dose-and time-dependent in the range used here. E, after treatment with 1 × and 2 × NETs for 48 h, the percentages of annexin V positive tumor cells increased compared with that of no NETs. F, representative images showed that the percentage of cells in the G0/G1 phase increased after a 48 h incubation, suggesting a contribution to the growth inhibitory eﬀects of NETs (*denote comparision with no-NETs). G, migration assay with plates and transwell chambers both indicated that NETs reduced migration. H, the results of migration and invasion assays, detected by optical density of stained crystal violet extracted from the cells invading the lower surface of the transwell filter, showed an obvious reduction in migration, and the results of T24 invasion were not significantly diﬀerent. Scale bar = 100 μm; *P < .05, **P < .01. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
extracellular fluorescence of NE and cit-H3 superimposed with DAPI, indicating co-localization, therefore confirming the presence of NETs (Fig. 1C). Following DNAse I or protease treatment, neutrophils could not produce NETs in response to BCG (Fig. 1D). Taken together, we concluded that BCG induced neutrophils to form NETs, which were mostly consisted of DNA and proteins.
3.2. BCG-activated tumor cells also elicit NETs formation, in which IL-8 and TNF-α from activated tumor cells both have an eﬀect
An increase of extracellular fluorescence in co-cultures of BCG-ac-tivated cancer cells with neutrophils suggested that NETosis was trig-gered (Fig. 2A). The major components of the extracellular lattices were chromatin and granule protein, as demonstrated by staining DNA, histones and NE (Fig. 2B). SEM revealed that dispersed filaments, characterized the morphology of NETs following supernatant stimula-tion, which was consistent with that from PMA stimulation (Fig. 2C). In addition, an abundance of NETs appeared in neutrophils activated by the supernatants from BCG-treated tumor cells, but few existed in neutrophils activated by the untreated ones (Fig. 2D). We then showed that cytokines in the supernatants of BCG-stimulated bladder cancer cells, including IL-8, TNF-α, and HMGB1, were significantly increased compared to cytokines from unstimulated cells (P < .05; Fig. 2E). The corresponding antibodies were used to neutralize these cytokines. In-hibition of IL-8 and TNF-α, but not HMGB1, markedly diminished NETs formation (P < .05; Fig. 2F).
3.3. NETs inhibit bladder cancer lines by preventing viability, the cell cycle and migration, as well as inducing apoptosis
To analyze the eﬀects of NETs on bladder tumor alone, in vitro prepared cell-free NETs were used in the following experiments. NETs induced significant cell floating and shrinking in a dose-dependent fashion (Fig. 3A). As detected by CCK-8, inhibition of cell proliferation by NETs treatment was dose- and time-dependent (P = .001; Fig. 3B). Moreover, the inhibition with NETs was largely abolished by protein inactivation, but not by DNA digestion (Fig. 3C). Additionally, BCG-NETs induced time- and dose-dependent cytotoxicity on tumor cells (P < .05and P < .001 respectively; Fig. 3D). In addition, NETs ex-posure dose-dependently induced apoptosis and G0/G1 phase arrest in BC cell (Fig. 3E–F), which partly accounted for the growth inhibitory eﬀects of NETs. Similar results were observed in the other cell line used; representative data from the T24 cell lines were shown. Given the dose-dependent cytotoxicity and growth inhibition, we used NETs at a low dose (0.5 ×) to analyze migration and invasion, and found an obvious reduction in migration through filters and on plates (P < .01; Fig. 3G–H), but no obvious decrease in invasion of T24 cells (P > .05; Fig. 3H).