br Fig Antitumor e ects of administration of BCG induced
Fig. 5. Antitumor eﬀects of administration of BCG-induced NETs.
A & B, comparison of diﬀerently treated subcutaneous tumors showed that the weights were significantly reduced in NETs and DNase-pretreated NETs group, but non-significantly in the boiling-inactivation group (* denote comparision to sham). C, the growth-time curve showed that tumor volume in the NETs group and the sham group increased with time. D, after administration of NETs, necrosis was evident with population by leukocytes. E, in situ cell death detection of the tumor tissue showed increased apoptosis or death cell in the NETs group than the sham group. F, after IHC staining, infiltrating inflammatory cells were observed in the tumor. CD3+ counting was higher in the NETs group than the sham or boiling-NETs group. G, the counting results of CD14+ were also higher in the NETs group; however, the increase was not apparent between the boiling-treated NETs group and the sham one. *P < .05, **P < .01.
concluded that BCG-induced NETs inhibited tumors via cytotoxicity, induction of cell-cycle arrest and apoptosis.
In the current study, BCG-activated tumor cells induced more NETs than non-activated ones. Stimulation with the supernatant of activated cells significantly increased neutrophil adhesion and NETs release, in-dicating an important role for cytokines. Certain cytokines have already been identified from epithelium or malignant cells treated by BCG [8,35]. NETs also formed in response to proinflammatory stimuli [29,36,37], including HMGB1, IL-8, and TNF-α that significantly in-creased after BCG stimulation in our study. IL-8 is a potent inducer of NETs, which is the strongest chemokine for PMNs, and involved in NETs formation mediated by activated endothelial cells [18,29]. HMGB1, a danger-associated molecular pattern mediator released from dying and activated cells, mediates neutrophil recruitment, reportedly promotes some proinflammatory cytokines  and induces NETs [21,37]. TNF-α, an important cytokine associated with tumors, is also an inducer of NETs [39,40]. This study demonstrated that IL-8 and TNF-
α mediated NETs induction by FG 4592 cancer cell, at least in part, following BCG stimulation.
A murine model of systemic infection demonstrated that micro-vascular NETs deposition and consequent trapping of systematically administered circulating lung carcinoma cells, increased metastasis . It cannot be overlooked, however, that systemic inflammation induces a multitude of eﬀects on numerous cell types, which may contribute to tumor adhesion and metastasis. In this study, we directly used cell- and bacteria -free NETs to rule out the roles of activated neutrophils and other active components. Addition of NETs resulted in an inhibition of migration and invasion of tumor cells.
CD4 is mainly expressed on the surface of Th cells. CD4+ T cells are of central importance in BCG therapy,  based on IFN-γ production or activation of CD8+ T and NK cells . Neutrophils are essential for T-cell traﬃcking to the bladder after BCG perfusion . Furthermore, it has been reported that NETs primed T cells,  function as remark-able DC activators and type I interferon inducers . PBMCs mainly consist of T cells, DCs, and monocytes, and the latter influenced NETs-mediated T cell activation . To confirm that NETs, but not neu-trophils, mediated this cellular immunity, cell-free NETs were used. We found that lower doses but not high doses of NETs led to proliferation of PBMCs and Th1 cytokine secretion. The negative correlation might be the result of activation and accompanying cytotoxicity of NETs . Also, we observed that NETs treatment increased CD4 expression in vitro, and CD3+ and CD14+ cells in tumors, indicating immune up-regulation. The local presence of Th1 cells, monocytes, and cytotoxic T-lymphocytes (CTLs) is generally associated with tumor regression and a favorable prognosis.
Previous study has shown that the toll-like receptor (TLR) 2 ligand, such as HP-NAP could inhibit the growth of BC by activating a cytotoxic Th1 response . Therefore, combined with the results of our study, we suggested that NETs might play a role in tumor progression by ac-tivating PBMCs, Th1 cytokine secretion, and CD4 expression. Never-theless, the activation cannot compensate for the inhibition of survival by high dose of NETs. Given that the density of neutrophils that accu-mulated in tissues cannot reach the levels in blood, the NETs in BCG-administrated tissues possibly mediated an anti-tumor eﬀect by in-hibiting tumors and activating the immune cascade. Our findings re-presented a novel mechanism of how NETs contribute to anti-tumor immunity.
Notably, neutrophils have either pro- or anti-tumor activity, de-pending on factors, such as cancer type and cytokine profile in the microenvironment [9,30]. It has been demonstrated that eﬃcient re-lease of myeloperoxidase (MPO) and NE, which are important compo-nents of NETs, is important for the antitumor eﬀect of neutrophils. At high concentrations, these factors are cytotoxic to tumor cells, but re-ducing the release might result in the conversion of neutrophil function from anti-tumor to pro-tumor [43,44]. Thus, we postulated that NETs might play diﬀerent predominant roles according to the diﬀerences of