Kaempferol blocks neutrophil extracellular traps formation and reduces tumour metastasis by inhibiting ROS‐PAD4 pathway

Abstract Kaempferol (kaem) is a dietary flavonoid found in a variety of fruits and vegetables. The inhibitory effects of kaem on primary tumour growth have been extensively investigated; however, its effects on tumour metastasis are largely unknown. In the present study, we found that kaem significantly suppresses both primary tumour growth and lung metastasis in mouse breast tumour model. Furthermore, decreased expression of citrullinated histone H3 (H3‐cit), a biomarker of neutrophil extracellular traps (NETs), had been founded in metastatic lung upon treated with kaem. The reduction of H3‐cit is not, however, due to the cytotoxicity of kaem on neutrophils since the frequency of CD11b+Ly6G+ neutrophils did not change in lung, tumour or blood in the presence of kaem. We then confirm the anti‐NETs effects of kaem in vitro by co‐culturing mouse neutrophils and kaem. Supplementing the neutrophils with GSK484, a potent NET inhibitor, totally abrogated the inhibitory effects of kaem on tumour metastasis while having little or no impact on primary tumour growth, indicating the specificity of kaem acting on NET formation and tumour metastasis. We also found that kaem suppressed ROS production in mouse bone‐marrow derived neutrophils. Supplementing with the ROS scavenger DPI abrogated kaem's effects on NET formation, suggesting the involvement of kaempferol in NADPH/ROS‐NETs signalling. Finally, we applied the kaem on NET‐deficient PAD4‐/‐ mice and found decreased primary tumour volume and weight but similar lung metastatic tumour with kaempferol treatment. Therefore, our findings reveal a novel mechanism of kaem in breast cancer development by targeting NETs induced tumour metastasis.

are many ways to manage this leading cause of mortality, such as surgery, radiation treatment and chemotherapy. However, none of these is a panacea for such a resilient disease. The high risk of recurrence is a major clinical manifestation and represents the principal cause of breast cancer-related death. 3 Chemotherapy is also limited by a narrow therapeutic index, significant toxicities and acquired resistance.
Kaempferol (kaem), which is a widely distributed dietary flavonoid, mostly found in fruits and plants such as apples, strawberries, tomato, broccoli, green tea and ginkgo leaves. [4][5][6][7] Current studies have demonstrated the bioactivity of kaem, including its anti-cancer, anti-inflammation, antioxidant, antidiabetic, neuroprotective and cardioprotective abilities. With regard to its anti-cancer activity, kaem regulates cancer cells through several different mechanisms, including but not limited to cell cycle arrest, promoting apoptosis and anti-angiogenesis properties. 8 However, the effects of kaem on tumour metastasis and cancer immunology remain unclear.
Neutrophils, which are a type of polymorphonuclear leukocyte, are well recognized as one of the major players during acute inflammation. 9,10 They are typically the first leukocytes to be recruited to an inflammatory site and are capable of eliminating pathogens via multiple mechanisms. Recent studies have revealed that the presence of a high neutrophil-to-lymphocyte ratio (NLR) has been associated with increased mortality in metastatic breast cancer, 10 suggesting a role of neutrophil in tumour metastasis. Specifically, the metastasis-supporting effects of neutrophil in breast cancer are related to the neutrophil extracellular traps (NETs). 11 The effects of NETs in tumour metastasis are confirmed by many studies, but the underlining mechanisms are varied. Some researchers think that the DNA mesh can trap circulating cancer cells at the site of dissemination and promote metastasis, 12 some think that NETs could activate toll-like signalling and MAPK signalling, which could increase tumour cell survival, 13 and other researchers think that NETs could increase local vascular permeability, 14 and aid cancer cell to extravasate from the blood vessel more easily.
In this study, we found kaem significantly inhibited breast cancer cell metastasis in vivo. These anti-metastatic effects were associated with inhibitory effects of kaem on neutrophil NET formation and dsDNA release. Treatment of the neutrophils with GSK484, a potent NETs inhibitor, totally abrogated the inhibitory effects by kaem, indicating the specificity of kaempferol acting on NETs.
Because the formation of NETs highly relies on the ROS production by neutrophil, we confirmed that kaem could suppress neutrophil ROS release, which contributes to its inhibitory effect on NET formation. Consistent with these in vitro results, deletion of NETs in transgenic PAD4 −/− mice abrogated the anti-metastasis effects of kaem in vivo. Therefore, our findings reveal a novel mechanism of kaempferol in breast cancer development by targeting NETs induced tumour metastasis.

| Drugs, reagents, antibodies and mice
Cancer cell line 4T-1 was obtained from the American Type Tissue Culture Collection (ATCC). The cell line was grown in Roswell Park Memorial Institute 1640 medium (RPMI) (Thermo Scientific, MA, USA), supplemented with 10% foetal bovine serum (FBS) (GIBCO), 2 mmol/L glutamine (GIBCO) and 100 U/mL penicillin and streptomycin (GIBCO). The cell line was cultured at 37°C in a humidified chamber with 5% CO2. Kaempferol (Sigma-Aldrich, USA) was dissolved in dimethyl sulphoxide (DMSO) to make a 10 mg/mL stock solution and stored at −20°C. The working dosage was freshly prepared in the basal medium with a final DMSO concentration of less than 0.1%. Anti-MPO antibody (AF3667) was sourced from R&D, anti-H3 citrulline (ab5103) and anti-H3 (ab18521) from Abcam. All of the animals used in the experimental procedures were approved by the Institutional Animal Care and Use Committee and Institutional Review Board of Hunan Provincial People's Hospital (No. 2014-07). All animals were kept under pathogen-free conditions at room temperature (21 to 25°C), with 12 hours light/ dark exposure. Food and water were offered ad libitum. Bodyweight was measured twice weekly as an indicator of overall animal health. All efforts were made to ameliorate the suffering of the animals during euthanasia by overdosing with CO 2 followed by cervical dislocation. Neutrophils were incubated with inhibitors 30 minutes before stimulation, as indicated. 20 μmol/L of the NADPH oxidase inhibitor, diphenyleneiodonium (DPI; Sigma-Aldrich, 300 260), was used. Further, phorbol 12-myristate 13-acetate (PMA, Sigma-Aldrich, P1585) was used at 10 nmol/L, NETs inhibitor GSK484 (Sigma-Aldrich, SML1658) was used at 100 μmol/L, exosome inhibitor GW4869 (Sigma-Aldrich, D1692) was used at 20 μmol/L, and the cells were incubated at 37°C. After 3 hours stimulation, cells were spun down and cell-free supernatant was collected. Supernatants were stored at −20°C until used.

| Cell proliferation
The exponentially growing cells (1 × 10 4 ) were plated in media in 48-well plates and treated with a series of concentrations of kaem dissolved in DMSO (final concentration ≤ 0.1%) after 24 hours of growth. Incubation was carried out at 37°C for 24 hours. Controls were formulated in DMSO vehicle at a concentration equal to that in drug-treated cells. The media was removed, and plates were washed with phosphate-buffered saline (PBS), air dried, stained with 0.3 mL methylene blue (2.5 g in 250 mL EtOH + 250 mL H2O), then left at room temperature (RT) for 2 hours. This was followed by washing with water, addition of 0.5 mL 1% Sarkosyl, and rotating at RT for 3 hours. 150 μL of this mixture was transferred to a 96-well plate, and the OD at 595 nm was determined using a plate reader. Cell viability assay was performed with three independent experiments.

| Wound-healing assays
Confluent cells were cultured with 25 µmol/L kaem for 24 hours and then wounded with a linear scratch by a disposable 200 μL micropipette tip. Wounded monolayers were then washed four times with RPMI cell culture medium to remove cell debris, and incubated in 10% FBS RPMI medium, with or without kaem, for up to 24 hours.
Cells migrated into the wound surface and the number of migrating cells were determined under an inverted microscope (Leica DMI 6000 B, IL, USA), by analysing five randomly chosen fields for each well. The percentage of inhibition was expressed using untreated wells at 100%. Three independent experiments were performed.

| Matrigel invasion assay
Cells were treated with 25 µmol/L kaem for 24 hours, and then har- Five independent experiments were performed.

| Lung metastasis in breast cancer mouse model
Female BALB/C mice (Animal Center of Hunan province, China), be-
Lung sections were stained for haematoxylin and eosin (H&E) to detect clear metastatic foci. Tissue sections were also stained for NETs markers such as MPO and citrullinated histone H3 (cit-H3). In brief, the sections were de-paraffinized, retrieved using 10 mmol/L citrate buffer at pH 6.0, and incubated at 95-100°C for 10 minutes. The sections were then incubated with protein block (10% FBS in PBS) followed by incubation with the primary antibody diluted in PBS, and kept at 4°C overnight. Then, the sections were washed with PBS and incubated with fluorescent secondary antibody. For analysis, five randomly selected areas from each section were photographed at 40× magnification, using Leica SP2 confocal microscope (Leica, IL, USA), and the number of labelled and unlabelled cells per area was determined by two independent observers using ImageJ software (NIH).

| Western blot analysis
Once proteins were extracted, they were quantified using Pierce BCA Protein Assay Kit (Thermo Scientific). The total cellular protein extracts were separated by SDS-PAGE and transferred to polyvinyldifluoride (PVDF) membranes (Millipore, MA, USA). The membranes were blocked with 5% (w/v) non-fat dry milk in TBST for 1 hour at room temperature, and then incubated overnight at 4°C with primary antibody.
Blots were washed three times in TBST buffer, followed by incubation for 1 hour at room temperature with the corresponding HRP-linked secondary antibodies. Specific proteins were visualized using enhanced chemiluminescence reagent (Thermo Scientific).

| Flow cytometry analysis
Organs were collected (n = 8 from each group) for flow analysis.
After extracting the tumour and lung tissues from mice, they were digested with collagenase II (1 mg/mL) and DNase (20 µg/mL) in RPMI media containing 5% FBS. Then, they were incubated at 37°C on a shaker for 1 hour and disseminated into single cells, filtered through a 40-micron mesh, and spun at 500g for 15 minutes.

| SYTOX Green plate reader assay for NETosis analysis
Neutrophils were isolated from mouse bone marrow using EasySep Mouse Neutrophil Enrichment Kit (STEMCELL Technologies Inc., MA, USA). NETs were generated using the protocol described earlier. 15 Briefly, Neutrophils were stimulated with 10 nmol/L PMA and incubated on a 25 mm flat tissue culture dish for 4 hours. After 4 hours, the media was gently aspirated and discarded, and the bottom layer was collected by pipetting 15 mL cold PBS (no Ca 2+ and Mg 2+ ). The suspensions were then centrifuged at 450 g for 10 minutes at 4°C to collect the cell-free NET-rich supernatant. The supernatant was divided into 1.5 mL tubes and spun for at 18 000 g 10 minutes at 4°C. Supernatant was then discarded and the pellet was resuspended in 100 µL of PBS.
Concentration of the sample ranged between 140 and 180 ng/µL.

| Statistical analysis
Quantitative data were expressed as standard error of the mean (± SEM) and analysed through Student t test or ANOVA (depending on the number of comparative groups), followed by Bonferroni test for normal distribution. Those groups with non-normal distribution were analysed using the Mann-Whitney test or Kruskal-Wallis test. Statistical significant differences were considered at P-value < 0.05 using GraphPad Prism 6 software.

| Kaem inhibits breast tumour metastasis in mice
We first examined the effect of kaem on the primary tumour growth and metastasis of 4T1 cells in BALB/C mice using ectopic implantation of cancer cells in the mammary fat pad ( Figure 1A).
This tumour models allowed us determine the formation of both primary and secondary metastatic growth in breast tumours. For the primary tumour growth, dose dependent effects can be seen between 40 and 160 mg/kg kaem. Upon treating with 160 mg/ kg kaem, the primary tumour volume decreased from 1500 mm 3 to 800 mm 3 (Figure 1B-C) and no significant bodyweight loss can be seen. We also assessed the primary tumour and metastasis in the lung by H&E staining (Figure 1D, Figure S1). In the primary tu-

| Kaem inhibits neutrophil extracellular trap formation in vivo and in vitro
To address the discrepancy in the in vitro and in vivo anti-metastatic effects demonstrated by kaem, we decided to measure the neutrophils in the animal model, especially since neutrophils have recently emerged as important players in breast tumour microenvironments, particularly in metastasis. We collected samples of primary tumour, blood and metastatic lung from tumorous mice with and without kaem treatment (Figure 2A). Analysis revealed clear neutrophil populations in primary tumours, blood and metastatic lungs, but no significant difference in the neutrophil numbers between control mice and kaem-treated mice. These results suggest that the neutrophil survival or infiltration is not impaired by kaem.
We also measured other immune cells in the metastatic lungs, such as T cells, NK cells, B cells and monocytes. Our data showed that kaem did not affect infiltration of those immune cells in metastatic lungs ( Figure S3). However, the expression of DNA and citrullinated histone H3 (H3 cit), a biomarker of neutrophil extracellular trap formation, was suppressed by half in MPO + neutrophils in metastatic lungs upon kaem treatment ( Figure 2B and C), indicating that the kaem may contribute to regulation of neutrophil NET formation.
Since kaem treatment showed NETs inhibitory effect in metastatic lungs, which might contribute to its anti-metastasis effect, our next question was whether the effect is directly driven by kaem, or mediated by other cells or micro-environmental factors. Therefore, we isolated fresh neutrophils from mouse bone marrow and stimulated them with phorbol 12-myristate 13-acetate (PMA) to induce NET formation. Supplementing the mouse primary neutrophils with kaem could inhibit double-strand DNA release, a hallmark of the formation of NETs. The reversible PAD4 inhibitor, GSK484, which blocks the citrullination of PAD4 targeted proteins in neutrophils and inhibits the formation of NETs, also showed significant reduction of dsDNA release ( Figure 3A).
Moreover, kaem could not further reduce dsDNA release with GSK484 pre-treated neutrophils ( Figure 3A). In the functional kinetic analysis, in which the PAD4 was overexpressed in HEK293 cells, the expression of citrullinated proteins was significantly inhibited by the kaem treatment ( Figure S4), suggesting that kaem may inhibit NET formation through a PAD4-dependent signalling pathway. We further confirmed our hypothesis by performing immunofluorescence assay and Western blot ( Figure 3B-D).

| Neutrophil extracellular traps promote breast cancer cell migration
Although recent studies indicate that neutrophils or NETs promote tumour metastasis, our understanding of these mechanisms is still limited. Here, we tested the effects of NETs on tumour cell migration in vitro by using co-culture models with neutrophil and breast cancer cell lines 4T1. We chose two different co-culture models, contact-model and separate-model ( Figure 4A-D). Enhancement of metastatic potency of breast cancer cells by neutrophils was Data presented as mean ± SEM of three independent experiments. *P < 0.05, compared with control group; **P < 0.01, compared with control; ***P < 0.001, compared with control ( Figure 4D). In addition, we performed a transwell assay using the isolated NETs-4T1 cell co-culture model. It showed that NETs dramatically increased the migration of 4T1 cell ( Figure S5). Inhibitory effects of DNase I on p-p38 and p-AKT signalling were also confirmed by Western blot, indicating that the NETs could promote p38 and AKT pathway signalling for cell survival and metastasis ( Figure 4E). Interestingly, a significant increase in NETs production in response to fMLP was detected in the neutrophils of patients who developed invasive breast cancer, when compared with age/gendermatched healthy volunteers ( Figure 4F).

| Kaem inhibits neutrophil extracellular traps formation through ROS signalling
NADPH oxidase derived reactive oxygen species (ROS) is indispensable for the process of NET formation. 16 Since kaem is a polyphenol antioxidant compound, 17 it is worthwhile to test its effect on ROS production in neutrophils and NET formation.
By using the classical PMA-induced ROS production model, we found that kaem could inhibit ROS production and dsDNA release in active neutrophils ( Figure 5A). However, kaem could not further lower the dsDNA release if the cells were pre-treated with Diphenyleneiodonium (DPI), an NADPH inhibitor ( Figure 5B), suggesting that the inhibitory effect of kaem on NET formation is mediated by ROS inhibition.

| Inhibition of NET formation in vivo fails to affect primary tumour growth, but abrogates the antimetastatic effect by kaem
To verify the anti-NETs effect of kaem in vivo, we use peptidyl arginine deiminase type IV (PAD4)-deficient mice 18 ( Figure 6A).
Compared to vehicle-treated 4T1 tumour bearing PAD4 −/− mice, kaem treatment showed 50% reduction of primary tumour F I G U R E 4 NETs promote breast cancer cell survival and metastasis. Primary mouse neutrophils and 4T-1 cells were co-cultured at opposite sides of transwell permeable supports (referred to as contact model). Neutrophils were grown on transwell permeable supports and 4T-1 cell were grown on the bottom of a separate well, the gap between the two cell layers being 0.8 mm (referred as separate model). MMP2 and MMP9 mRNA levels were measured by 10 nmol/L PMA treated for 2 h (A), pre-treated with 20 µmol/L exosome inhibitor GW4869 for 1 h, and then added 10 nmol/L PMA for additional 2 h (B), followed by pre-treatment with 1000 U DNAse I for 1 h and addition of 10 nmol/L PMA for another 2 h (C), and 4T-1 cells co-cultured with isolated NETs (300 mg/dL of protein) (D). (E) NETs production of neutrophils from health and breast cancer patients. (F) Western blot of p38 and AKT pathway for samples from 5A-C. Data presented as mean ± SEM of three independent experiments. *P < 0.05, compared with control group; **P < 0.01, compared with control; ***P < 0.001, compared with control volume (Figure 6B), and 50% reduction of tumour weight ( Figure 6C). However, there is no significant difference of the metastatic tumour in the lung between vehicle and kaem treated mice ( Figure 6D-E). These results suggested that kaem could inhibit both primary tumour growth and tumour metastasis by distinct mechanisms. The possible mechanism through which kaem inhibits tumour metastasis is mainly through the ROS-PAD4 signalling. Neutrophils, however, have only recently emerged as important players, particularly in metastasis. 19 As highlighted in other recent studies, neutrophils can influence tumour metastasis via their ability to produce and release a wide variety of proteins into the extracellular environment. However, in our co-culture model, we did not see any significant reduction of metastatic ability in mouse breast cancer cell lines if neutrophils and tumour cells were cultured separately in two chambers, indicating the influence between cells is limited by distance parameters. We also tested the role of exosomes in our co-culture model, because as a newly discovered cell-cell communication mechanism, neutrophil-derived exosome can modulate cell migration, cell differentiation and other aspects of cell-to-cell communication. However, we still could not see significant reduction of metastatic ability in mouse breast cancer cell lines after GW4869 treatment. We do see a dramatic reduction by using NETs inhibitor DNase I treatment, which means NETs could promote cancer cell metastasis. This observation was also confirmed using co-cultured purified NETs and cancer cell lines. The stimulatory effects of NETs in tumour metastasis have been confirmed by many studies, but the underlining mechanisms are varied. Some researchers think that the DNA mesh can trap circulating cancer cells at the site of dissemination and promote metastasis, 12 some think that NETs could activate toll-like signalling and MAPK signalling, which could increase tumour cell survival, 13 while some other researchers think that NETs could increase local vascular permeability, 14 and aid cancer cell to extravasate across blood vessels more easily. Due to the limitations of our co-culture model, we could not get the possibility of the NETs effects on circulation cancer cells and vascular permeability, but we confirmed that NETs could activate MAPK signalling and MMP signalling.
Although targeting NETs to prevent metastasis has been proposed by many papers and showed promising effects, the clinical or pre-clinical medicine is still limited. Some newly discovered technology, such as DNase I-coated nanoparticles, 11 and newly discovered targets, such as PI3K-gamma, 20 propose a promising therapeutic strategy to block NET formation in vivo. However, their safety and efficiency need to be evaluated in the long run. In our study, we found kaem works well against breast tumour metastasis. Many studies focus on the anti-proliferation or promote apoptosis effects of kaem on tumour cells. Here, we explore the anti-metastatic effects of this popular flavonoid. In vitro results demonstrated that kaem inhibited NETs release in the isolated neutrophils with a significant reduction of citrullinated histone H3, a hallmark of NET formation. We also showed that kaem cannot further reduce NET formation with PAD4 blocker pre-treatment, suggesting that kaem may inhibit NET formation through a PAD4dependent signalling pathway. This assumption was confirmed F I G U R E 7 An overview of the mechanism of action of kaempferol in tumour metastasis by PAD4's functional kinetic analysis (measuring citrullination directly) and validated in PAD4-deficient mice. However, we still cannot rule out the possibility of kaem reducing NET formation through other downstream, such as Syk, Src kinases or PI3K IAδ, 21 which needs further investigation. Therefore, there is great future potential for kaem to work along or in combination with other reagents in order to treat metastatic tumours (Figure 7).

ACK N OWLED G EM ENT
This work was supported by the Hunan Provincial Nature Science Foundation of China (No. 2018JJ6105).

CO N FLI C T O F I NTE R E S T
All authors read and approved the final version of the manuscript, and the authors declare that they have no competing interests. Zeng and Chao-jie Zhang contributed to draft the work. All authors contributed important intellectual content for the overall work and take responsibility for the honesty and accuracy of the present study.

E TH I C A L A PPROVA L
The study was approved by the ethics committee of Hunan Provincial People's Hospital (No. 2014-07). All patients provided written informed consent before taking part in the study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.