Neutrophil extracellular DNA traps promote pancreatic cancer cells migration and invasion by activating EGFR/ERK pathway

Abstract Neutrophil extracellular DNA traps (NETs) are newly discovered forms of activated neutrophils. Increasing researches have shown that NETs play important roles in cancer progression. Our previous study has proved that tumour‐infiltrating NETs could predict postsurgical survival in patients with pancreatic ductal adenocarcinoma (PDAC). However, the roles of NETs on the progression of pancreatic cancer are unknown. Here, we investigated the effects of NETs on pancreatic cancer cells. Results showed that both PDAC patients’ and normal individuals’ neutrophils‐derived NETs could promote migration and invasion of pancreatic cancer cells with epithelial‐mesenchymal transition. Further, study confirmed that EGFR/ERK pathway played an important role in this progression. The addition of neutralizing antibodies for IL‐1β could effectively block the activation of EGFR/ERK companied with reduction of EMT, migration and invasion. Taken together, NETs facilitated EMT, migration and invasion via IL‐1β/EGFR/ERK pathway in pancreatic cancer cells. Our study suggests that NETs may provide promising therapeutic targets for pancreatic cancer.


| INTRODUC TI ON
Patients with pancreatic ductal adenocarcinoma (PDAC) have an extremely poor prognosis, with a 5-year overall survival rate of less than 5.0%. 1 Pancreatic cancer has a higher mortality rate due the high incidence of recurrence and metastases dissemination. 2 Thus, figuring out the mechanisms regulating pancreatic metastasis is essential to treat high-risk patients.
Carcinogenesis is closely linked to a dysfunctional immune system. 3 The interplay between cancer and the immune system may influence many aspects of cancer progression. 4 Neutrophils are generally the most abundant immune cell population and involved in the development of cancer. 5 Recent studies suggest that high peripheral blood neutrophil-to-lymphocyte ratio is an indicator of poor survival in patients with pancreatic cancer. 6 Moreover, tumour-infiltrating neutrophils are frequently observed in patients with pancreatic cancer and are associated with reduced survival. 7 Under specific stimulatory conditions such as infection and reactive oxygen species, neutrophils release chromatin structures which are decorated with specific cytoplasmic and granular proteins. These structures then form DNA meshes; special forms of neutrophils called neutrophil extracellular DNA traps (NETs).
As a part of innate immunity in humans, neutrophils release decondensed chromatin networks to capture and disarm bacteria and fungi via NETosis. Citrullination of histone H3 (Cit-H3) by peptidylarginine deiminase 4 is an important step in NETosis. 8 In addition, the activity of myeloperoxidase (MPO) is critical in the process of DNA extrusion. Both Cit-H3 and MPO serve as markers for formation of NETs. 9 NETs play important roles in the innate immune response, pathologic alteration and development of cancer progression. 10 It has been reported that breast cancer cells can induce metastasis-supportive NETs. 11 NETs could promote cancer metastasis by sequestering and entrapping circulating cancer cells through interactions induced by b1-integrin. 12 NETs formation is conducive for implantation of ovarian cancer cells, and blockade of NETs formation could prevent omental metastasis. 13 NETs could act as an adhesion substrate for different tumour cells. 14 NETs also act as a drug target to counteract chronic and acute inflammation. 15 It has also been reported that NETs could promote EMT 16 and enhance metastasis through CCDC25 17 in breast cancer.
The high mortality of PDAC is attributed to its aggressive biological phenotype, which is characterized by frequent invasion and metastasis. 18 80% of pancreatic cancer patients are diagnosed with metastasis, and about 50% of patients suffer liver metastasis within 2 years after receiving surgery and adjuvant therapy. Evolutionary biological process called epithelial-to-mesenchymal transition (EMT) contributes to pancreatic metastatic dissemination. 19 During the process of pancreatic cancer EMT, epithelial marker proteins, such as E-cadherin 20 and ZO-1, 21 decline. And the expression levels of mesenchymal markers N-cadherin 22 and vimentin 23 increase. EMT is regulated by complex regulatory networks in pancreatic cancer by EMT-related transcription factors, especially SNAIL (zinc finger protein SNAIL), 20 ZEB (Zinc finger E-box binding homeobox 1) 24 and Twist (twist basic helix-loop-helix transcription factor 1). 25 Matrix metalloprotein such as MMP2 and MMP16 also plays important roles in pancreatic EMT. 26 Our previous studies have shown that NETs are associated with poor prognosis and high frequency of metastasis in patients with PDAC. 27,28 The specific mechanism of NETs promoting the invasion and metastasis of pancreatic cancer has not been clearly studied.
Our study shows that NETs promote PDAC migration, invasion and EMT through IL-1β/EGFR/ERK pathway. And inhibition of IL-1β/ EGFR/ERK pathway could effectively inhibit NETs-induced migration, invasion and EMT of pancreatic cancer. Island, NY, USA). And, MIA PaCa2 was cultured in DMEM containing 10% FBS (Gibco, Grand Island, NY, USA). All medium were supplemented with 100 μg/mL streptomycin and 100 U/mL penicillin.

| Cell culture
All pancreatic cancer cell lines were cultured in a humidified atmosphere containing 5% CO 2 at the temperature of 37°C.

| Patients and specimens
Patients with PDAC were subjected to the following inclusion condi- The associated permit number is 050432-4-1212.
Immunohistochemical staining was evaluated by two independent pathologists who were blinded to the clinical data. The intensity of the staining was graded as 0, 1 or 2. The percentage of positive cells was graded as 0 (0%), 1 (≤1%), 2 (≤5%), 3 (≤10%), 4 (≤20%) or 5 (>20%). The two grades were added together. A total score of 0 to 1 was considered as low expression. A total score of 2-5 was considered as a median expression. A total score of 6 to 7 was considered as high expression.

| Conditioned medium (CM)
We extracted neutrophils from the blood, inoculated them in 5% RPMI medium which was made by supplementing RPMI media with 5% foetal bovine serum (FBS) at the concentration of 5 × 10 6 neutrophils/ml. Stimulate neutrophils with 5 nM of PMA for 8hr at 37°C 5% CO2. This will allow NETosis. Collected conditioned medium (CM) was centrifuged at 2000rp/min and filtered through 0.2um filter to remove the DNA meshes of NETs. Collected conditioned medium (CM) was concentrated using an Amicon Ultra Centrifugal Filter device (Merck Millipore) with a molecular weight cut-off of 10 kDa.
10 ml conditioned media were concentrated to 0.5 ml. We stored the CM at -80°C until use. The CM was added into the culture medium at a 1:10 ratio.

| Drugs and antibodies
The antibodies used were anti-phospho-EGFR1068 (Cell Signaling

| Real-time PCR
Total RNA was extracted using the TriPure Isolation reagent (Roche).
Complementary DNA was obtained using the M-MLV Reverse Transcriptase Synthesis kit (Promega). qPCR was subsequently carried out using the Power SYBR Green PCR Master Mix kit (Applied Biosystems). Relative quantities were calculated using the ΔΔCt method and the endogenous reference β-actin. Each experiment was performed in triplicate.
Primer sequences for qPCR.

Gene Sequence
Snail Up

| Western blot
Cells were collected and lysed in NETN 150 lysis buffer. The quantities of proteins were identified using the Bradford assay. 20 μg protein was then separated by SDS-PAGE and transferred to a nitrocellulose membrane (Axygen, Tewksbury, MA, USA). The membrane was incubated with 5% non-fat milk for 1 h at room temperature and then with the indicated primary antibodies (1:1000) at 4°C for 12 h. The membrane was then incubated with HRP-conjugated secondary antibodies (1:1000). We used the Millipore Immobilon Western Chemiluminescent HRP Substrate kit to detect the protein expression.

| Cell proliferation assay
Cell proliferation was determined using the cell counting kit 8 (Dojindo Molecular Technologies, Kumamoto, Japan). 29 Each experiment was repeated three times with triplicate samples.

| Statistical analysis
All statistical analyses were carried out using GraphPad Prism 6 software. Each result is shown as the mean ±SD of three independent experiments (n = 3). The statistical significance of comparisons among multiple groups was assessed by one-way analysis of variance, followed by Bonferroni analysis, with P <.05 considered statistically significant. Correlation between two groups was assessed by Pearson's correlation analysis. P value less than 0.05 was considered statistically significant (*: 0.05, **: < 0.01 and ***: < 0.001).

| NETs promote migration and invasion of pancreatic cancer cells
To investigate the influence of NETs on pancreatic cancer cells, we isolated neutrophils from patient blood and induced the formation  Figure 1C-F, Figure S1A, B).
As previous studies reported that neutrophils could be divided into anti-tumourigenic N1 phenotype and pro-tumourigenic N2 phenotype basing on their function. Until now, no specific marker has been identified to distinguish the N1/N2 subgroups in clinical research and diagnosis. 27 (Figure 2A-D). The same findings were also verified in PANC1 cells ( Figure S2A and B). There was no difference in the function of NETs derived from PDAC patients or normal donors. The specific mechanism is important for pancreatic cancer research and needs further study.

| NETs induce EMT in pancreatic cancer cells
One of important process in cancer metastasis is EMT, by which cancer cells lose their epithelial characteristics and exhibit mesenchymallike features with high migratory and invasive capacities. 30 To determine whether NETs could influence EMT, we treated

| Tumour migration, invasion and EMT induced by NETs are dependent on EGFR/ ERK signalling
To determine the mechanical link between NETs and EMT in pancreatic cancer cells, we investigated several important EMT relative pathways such as EGFR, 31 WNT, 32 TGFβ, 33 AKT 34 and mTOR 35 pathway.
The results showed that NETs activated EGFR/ERK in both BxPC3 and MIA PaCa2, whereas the inhibition of NETs blocked the activation ( Figure 4A). While other EMT-related pathways such as WNT, PI3K and TGFβ were not affected by NETs in pancreatic cancer cells. Western blot results showed that NETs upregulated EGFR/ERK pathway specifically. Importantly, the inhibition of both EGFR (by the EGFR inhibitor erlotinib) and ERK (by the ERK inhibitor SCH772984) blocked the induction of EMT-related genes by NETs ( Figure 4B). All the results collectively suggested that EGFR/ERK pathway was important in NETs-induced EMT.
To determine whether NETs mediated migration and invasion of pancreatic cancer cells are EGFR/ERK dependent, we examined the effects of EGFR and ERK inhibitors on pancreatic cancer cells after treatment with CM-NETs. The results of wound healing assay and transwell assay showed that EGFR/ERK inhibition significantly reduced NETs-induced pancreatic cancer cell migration and invasion without influencing proliferation ( Figure 4C-F, Figure S3A and B).
Taken together, these results showed that NETs promoted EMT, migration and invasion via the EGFR/ERK pathway.

| IL-1β is the mediator of NETs-induced migration, invasion and EMT
Next, we examined the mechanism by which NETs could activate the EGFR pathway. It has been shown that various kinds of cytokines are NETs was inhibited ( Figure 5A). The data showed that NETs released more than 10 times of IL-1β than neutrophils from PDAC patients.
ELISA results also confirmed that almost all IL-1β were from the secretion of NETs ( Figure 5B). It has been reported that IL-1β plays important roles in EGFR activation, such as IL-1β transactivates EGFR via the CXCL1-CXCR2 axis. 41 It has also been shown that IL-1β plays Taken together, these data suggest that IL-1β is responsible for NETs-induced pancreatic cancer cell mobility and EMT.

| NETs correlate with EGFR activation and EMT in vivo
To  Another study has suggested that NETs could promote both the invasion and expansion of cancer cells by concentrating proteases and protecting them from endogenous inhibitors. 11  Real-time PCR was used to assess the expression of the EMT-related genes in BxPC3 (A) and MIA PaCa2 (B) cells after the specified treatment. C and D, Vimentin (green) and N-cadherin (red) were detected by immunofluorescence in BxPC3(C) and MIA PaCa2 (D) after the specified treatment. Nuclei were stained with DAPI (blue). E and F, Western blot analysis of the EMT-related proteins in BxPC3 (E) and MIA PaCa2 (F) after the specified treatment. PMA was used to induce NETs formation. Sivelestat (40 nM) was used to inhibit the formation of NETs

CO N FLI C T O F I NTE R E S T
The authors confirmed that there was no conflict of interest. Xianjun Yu: Supervision (equal).

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.

O RCI D
Wei Jin https://orcid.org/0000-0002-3639-1547 F I G U R E 6 NETs promote EGFR-dependent EMT in vivo. A. Ventral flanks of female BALB/c nude mice were injected with 2.5 × 10 6 MIA PaCa2 cells subcutaneously. When the tumours reached approximately 4 × 4 mm in size, tumour-bearing mice were randomly assigned to different experimental groups (n = 4 per group) and treated with different types of conditioned medium. We seeded neutrophils at a density of 5 × 10 6 /ml with or without PMA and collected the conditioned medium 8 h later. Collected conditioned medium (CM) was concentrated using an Amicon Ultra Centrifugal Filter device (Merck Millipore) with a molecular weight cut-off of 10 kDa. 10 ml conditioned media were concentrated to 0.5 mL. Intratumoural injection of 500 µL concentrated CM was performed every day. The volume was calculated according to the formula: V = ½ × length × width 2 . On day 24, the tumours were excised and photographed. The data were shown as mean with SD of four mice in each group. B, Immunohistochemistry staining of nude mice tumour tissue sections for EGFR, p-EGFR, N-cadherin and vimentin were shown. C, Immunohistochemistry staining of citH3, p-EGFR E-cadherin in PDAC patients. D, Relationship between CitH3, p-EGFR and E-cadherin expression in 150 PDAC patients' tissues