Tumor Biol. DOI 10.1007/s13277-015-3484-1

RESEARCH ARTICLE

Intratumoral neutrophil granulocytes contribute to epithelial-mesenchymal transition in lung adenocarcinoma cells Pingping Hu 1,2 & Meixiao Shen 1,3 & Ping Zhang 1 & Chunlong Zheng 1 & Zhaofei Pang 1 & Linhai Zhu 1 & Jiajun Du 1,4

Received: 2 December 2014 / Accepted: 21 April 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015

Abstract We previously demonstrated that haemoptysis as a prognostic factor in lung adenocarcinoma and haemoptysis was associated with severe vascular invasion and high circulating white blood cell count. Epithelial-mesenchymal transition (EMT) plays an important role in tumor invasion. We hypothesized there was some relationship between tumorassociated inflammatory cells, tumor invasion, EMT, and haemoptysis. Immunohistochemistry (IHC) was used to detect CD66b and E-cadherin expression in tumor tissue. By coculture tumor cells with polymorphonuclear neutrophils (PMNs), the expressions of EMT markers were assessed by western blotting. TGF-β1 concentrations in the supernatant and the migration activities of tumor cells were performed by ELISA and migration assays. Intratumoral CD66b+ PMN expression was negatively associated with E-cadherin expression. Haemoptysis was significantly associated with neutrophil infiltration (OR = 4.25, 95 % CI 1.246–14.502). Neutrophils promoted EMT of tumor cells in vitro and Pingping Hu and Meixiao Shen contributed equally to this work. * Jiajun Du [email protected] 1

Institute of Oncology, Shandong Provincial Hospital Affiliated to Shandong University, Shandong University, 324 Jingwu Road, Jinan 250021, People’s Republic of China

2

Department of Radiation Oncology, Qianfoshan Hospital Affiliated to Shandong University, Jinan, Shandong 250014, People’s Republic of China

3

Blood Purification Center, Hainan General Hospital, 19 Xiuhua Road, Xiuying District, Haikou, Hainan Province 570311, People’s Republic of China

4

Department of Thoracic Surgery, Shandong Provincial Hospital Affiliated to Shandong University, Shandong University, 324 Jingwu Road, Jinan 250021, People’s Republic of China

enhanced the migration activity of tumor cells. In addition, TGF-β1 was up-regulated and Smad4 translocated into nucleus, indicating that TGF-β/Smad signaling pathway was initiated during the process. We indicated that lung adenocarcinoma with haemoptysis was associated with more PMN infiltration and PMNs promoted EMT, partly via TGF-β/Smad signal pathway. This may provide mechanistic reasons for why haemoptysis was associated with poor outcome in lung adenocarcinoma. Keywords Tumor-associated neutrophil . Lung adenocarcinoma . Epithelial-mesenchymal transition

Introduction Tumor microenvironment consists of distinct cell types, including tumor cells, stromal cells, blood vessels, and infiltrating inflammatory cells [1]. Polymorphonuclear neutrophils (PMNs) belong to the latter and influence the tumor microenvironment by producing cytokines and chemokines, which alter the immune response [2]. Recently, increasing evidences have shown that PMNs are correlated inversely with patient survival in many cancers, either in tumor tissue or blood [3]. In our prior work, we found that haemoptysis, an unfavorable prognostic factor in operable lung adenocarcinoma, was associated with high level of circulating white blood cells (WBC) before operation, and vascular invasion rather than tumor necrosis or angiogenesis could be the vital mechanism of nonlife threatening haemoptysis in lung adenocarcinoma [4]. Interestingly, subsequent analyses revealed that there was an increase in the number of peripheral neutrophils in haemoptysis patients compared to non-haemoptysis patients. In a recent study, Fridlender demonstrated that the presence of TGF-β induced the formation of a pro-tumorigenic (N2)

Tumor Biol.

phenotype capable of supporting tumor growth and suppressing the antitumor immune response; in contrast, TGF-β blockade induced the accumulation of antitumorigenic neutrophils (N1) which activated CD8+ T cells and were more cytotoxic against tumor cells in vitro [5]. N2 phenotype neutrophils have been recognized as key mediators in tumor initiation, growth, progression, and immunomodulation [6]. Thomas et al. found that PMNs induced epithelialmesenchymal transition (EMT) of pancreatic tumor cells and promoted their motility [7]. EMT, an important developmental program, describes the transition of cellular phenotype from an epithelial to mesenchymal state. In this process, cytoskeletal components and extracellular matrix (ECM) are reorganized to promote a more invasive phenotype [8]. As a result, the expressions of the epithelial phenotype markers are lost during EMT occurs, such as E-cadherin and cytokeratin. In parallel, genes encoding mesenchymal proteins are increased, such as N-cadherin, vimentin, fibronectin, and nucleus β-catenin [9]. EMT has been shown to play important roles in tumor progression and metastasis [9, 10]. Several molecular pathways that mediate EMT in cancer cells have been identified, such as TGF-β, Ras, Notch, and Wnt/ β-catenin signaling pathways [11]. Among these, TGF-β/Smad signal pathway has received much attention in the process of EMT. Based on the views of previous studies, we hypothesized there might be some relationship between tumor-associated inflammatory cells, tumor invasion, EMT, and haemoptysis. We presumed that PMNs played crucial roles in promoting poor outcome of lung adenocarcinoma patients with haemoptysis, partly via inducing EMT in tumor cells. Therefore, we performed the study to address the following questions: (i) if there was a relationship between the intratumoral PMN infiltration and E-cadherin expression in tissue, (ii) if PMNs had an influence on the EMT of the lung adenocarcinoma cells in vitro, and (iii) if TGF-β/Smad signaling pathway was involved in the interactions of tumor cells and PMNs.

Materials and methods

University, and written informed consent was obtained by participants for their clinical records to be used in this study. The baseline WBC and NEUT (neutrophils) count were collected from routine blood tests performed 1–3 days before surgery. Overall survival (OS) was measured from date of surgery to either death, the last follow-up, or lost to follow-up. Immunohistochemistry and its score To explore the relationship between hemoptysis and PMN infiltration, we consecutively selected the cancer specimens from 24 haemoptysis and 24 nonhaemoptysis patients for immunohistochemistry (IHC) analyses. Formalin-fixed paraffin embedded surgical specimens were used for IHC. The detail procedures of IHC were described previously [12]. Briefly, the tissue blocks were sectioned at 4 μm and mounted on glass slides. Primary antibodies were against CD66b (1:600 dilution, BD, San Jose, CA, USA) and Ecadherin (Working solution, Zhongshan Biotechnology, Beijing, China). Diamino-benzidine (DAB) was used to detect the level of colorimetry. Nuclei were counterstained using hematoxylin. Blinded scoring was performed by two observers independently according to the immunoreactive score (IRS). The degree of Ecadherin expression was scored as intensity of staining multiplied by the percentage of positive cells. The intensity was defined as 0, negative; 1, weak; 2, moderate; and 3, strong. The percentage of positive cells was determined as 0, negative; 1, no more than 10 % positive cells; 2, 11 to 50 % positive cells; 3, 51 to 80 % positive cells; and 4, over 80 % positive cells. An IRS of 1–4 was identified as weak, IRS of 5–8 represented medium, and IRS of 9–12 was defined as strong [13]. The staining of intratumoral CD66b was defined as negative, low expression, and high expression based on tertiles; absent was defined as negative expression while present was defined as low and high expressions. The positive neutrophils was counted by eyes in three typical high-power fields (HPF, 400×), and the average numbers were calculated for further analyses.

Patients Isolation and culture of PMNs Between January 2006 and December 2011, 666 lung adenocarcinoma patients undergoing complete resection were consecutively selected from the Shandong Provincial Hospital Affiliated to Shandong University, including 149 haemoptysis and 517 non-haemoptysis patients (The characters of patients have been described previously [4]). Patients treated with preoperative chemoradiotherapy before surgery were excluded. All studies were approved by the Ethical Committee of Shandong Provincial Hospital affiliated to Shandong

PMNs were isolated from the venous blood of patients with lung adenocarcinoma by using Polymorphprep (Axis-Shield PoC, Oslo, Norway). The isolated PMNs were cultured in RPMI-1640 supplemented with 10 % fetal calf serum and used within 1 h. Morphologically, neutrophil purity was typically 90–95 %. The trypan blue exclusion test disclosed a cell viability of ≥95 % and ≥85 % after 1 h and at the end of the experiments, respectively.

Tumor Biol.

Cell lines and co-culture of cancer cells with PMNs Lung adenocarcinoma cell lines A549 and SPC-A1 were obtained from the Chinese Academy of Science Cell Bank. They were maintained in RPMI-1640 supplemented with 10 % fetal bovine serum and cultured at 12-well culture plates (Costar, Corning, NY, USA). The culture medium was replaced with serum-free RPMI-1640 when cancer cells grew to approximately 80 % confluence. As direct co-culture of cancer cells with PMNs group, PMNs were resuspended in serum-free RPMI-1640 (3×106/ml) and added to each well. As the indirect co-culture group, 12-well transwell inserts (membrane pore size: 0.4 μm, Costar, Corning, NY, USA) were used to separate cancer cells (at lower chambers) from PMNs (at upper chambers). As control (defined as Bmonoculture^), cancer cells were cultured in the culture plate alone in equal amount of serum-free RPMI-1640. The protein samples were collected for western blotting (WB) analyses after 3 h. In order to isolate subcellular compartments of cancer cells and collect the supernatants for further studies, A549 and SPC-A1 were seeded at 100-mm culture dishes (Costar, Corning, NY, USA). PMNs were added into the dish when cancer cells have grown to approximately 80 % confluence, which was defined as co-culture. Cancer cells were monocultured in the dish as control. The tumor cells and supernatants were collected after 24 h for nucleus protein analyses and ELISA, respectively. After direct co-culture, cancer cells were washed 3 times with phosphate buffered saline to wash off the non-adherent suspension PMNs, and then the adherent tumor cells were collected for further WB or migration assay.

Western blotting (WB) Protein samples of tumor cells were dissolved in NuPAGE LDS Sample Buffer (1.5×, Invitrogen, Carlsbad, CA) while subcellular compartments of cells (cytoplasm, nucleus) were lysed with an isolation kit (BestBio, Shanghai, China). Protein was quantified by using NanoDrop 2000 (Thermo Scientific, Wilmington, USA), and 50 μg of total protein or 20 μg of cytoplasm/nucleus protein was loaded to each well. Samples were analyzed by SDS–PAGE with a 10 % separation gel and transferred to nitrocellulose (NC) membranes at room temperature for 2 h. Prior to incubation with the first antibodies at 4 °C overnight, the NC membranes were blocked with 5 % BSA (Seasky Bio Technology, Beijing, China). Subsequently, the NC membranes were washed and incubated with the secondary antibodies for 1 h at room temperature. The blot was detected with Western Bright ECL WB HRP Substrate Trial kit (APG Bio, Shanghai, China). The densitometry was performed using ImageJ Software (NIH, Bethesda, MD, USA) by calculating the integrated density of each band and

normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Antibodies of WB Polyclonal rabbit anti-human E-cadherin (CDH1) antibody was obtained from Abgent (1:1500, Abgent, San Diego, CA, USA). Monoclonal mouse Smad4 antibody (1:1000), monoclonal mouse GAPDH antibody (1:2000), and polyclonal rabbit N-cadherin antibody (1:200) were from Santa Cruz Biotechnology (Santa Cruz, Dallas, Texas, USA). Rabbit polyclonal β-catenin antibody (1:1000) and rabbit monoclonal vimentin antibody were obtained from Cell Signaling Technology (1:1000, Cell Signaling Technology, Inc., Danvers, MA, USA). The secondary antibodies (HRP-goat anti-rabbit, HRP-goat anti-mouse) were from Zhongshan Biotechnology (1:25000, Zhongshan Biotechnology, Beijing, China). Migration assays Migration of tumor cells was assessed with the 24-well transwell inserts (membrane pore size: 8.0 um, Costar, Corning, NY, USA) according to the manufacturer’s instructions. Tumor cells were cultured in three groups for 24 h (see above Bco-culture of cancer cells with PMN^) and digested by 0.25 % Trypsin-EDTA; then 5×105 tumor cells were added to the upper chambers. The cells were allowed to migrate in a certain time (A549: 8 h, SPC-A1: 24 h). Then non-migrating cells on the upper surface of the membrane were removed by swabbing. After fixing and staining, the migrating cells were photographed at 100× magnification in 3 fields under microscope and quantification was performed with the image analysis software (Image Pro Plus 6.0; Media Cybernetics, Bethesda, MD, USA). The assay was performed in triplicate and each was repeated three times. ELISA The supernatants of monoculture or co-culture were used to detect the concentrations of TGF-β1 by means of ELISA kits (Lichen Bio, Shanghai, China). Concentrations of the culture supernatants were determined according to the manufacturer’s instructions. This trial was performed in triplicate and repeated three times. Statistical analysis Statistical analysis was done using SPSS software program (version 20.0; SPSS Inc., Chicago, IL, USA). Correlations of haemoptysis with TAN were calculated using Pearson’s X2 test. Correlations of intratumoral neutrophils and clinicopathological features were tested with Spearman’s rank

Tumor Biol.

correlation coefficient (Spearman r). The protein expression levels and migration activity of each two groups under different culture conditions were analyzed using Wilcoxon signed ranks test. Mann–Whitney U test was used to compare the distribution of NEUT in haemoptysis and non-haemoptysis groups. p≤0.05 was considered statistically significant.

Results PMN infiltration and EMT in tissue of lung adenocarcinoma Our previous studies demonstrated that haemoptysis as a prognostic factor in lung adenocarcinoma after curative resection, vascular invasion could be the most important mechanism of haemoptysis, and haemoptysis was associated with high WBC count [4]. To explore the relationship between tumor-associated inflammatory cells, tumor invasion, and haemoptysis, we further collected the NEUT count of the 666 lung adenocarcinoma patients (including 149 haemoptysis and 517 non-haemoptysis patients) and demonstrated that there was also a relationship between haemoptysis and high NEUT count (p = 0.006, Fig. 1a). Then, by performing IHC in 48 selected patients, we found that there were significantly more infiltrated PMNs in haemoptysis group. Haemoptysis significantly correlated with the increased incidence of PMN infiltration in tumor (OR=4.25, 95%CI 1.246–14.502, haemoptysis vs. non-haemoptysis). The results of IHC demonstrated that intratumoral neutrophil expression was positively correlated with neutrophil count (correlation coefficient=0.446, p=0.002) and negatively correlated with E-cadherin expression (correlation coefficient=−0.471, p=0.001) (Figs. 1b and 2). The infiltration of CD66b+ neutrophil (200×, Fig. 2a, c) in lung adenocarcinoma is negatively associated with E-cadherin expression (B and D) in the same sample. Intratumoral neutrophils were absent (A), whereas the expression of E-cadherin was strong in the same sample (B); when intratumoral neutrophils were present (C), the expression of E-cadherin was weak in the same sample (D). PMNs induce EMT of tumor cells in vitro In the further study, a co-culture system to investigate the relationships between tumor cells and PMNs was established, as illustrated in methods. This system could simulate the reciprocal (bi-directional) interactions of tumor cells and neutrophils in the tumor microenvironment. We initially confirmed whether PMNs could induce EMT in vitro. The expression of EMT markers, including E-cadherin, N-cadherin, vimentin, and β-catenin (cytoplasm and nucleus), was detected in monoculture, indirect co-culture, and direct co-culture

after 3 h. In both A549 and SPC-A1 cell lines, we observed a significant decrease in E-cadherin expression upon stimulation by co-culture with PMNs; the decline was more obvious in direct co-culture group than indirect group (each two group p<0.05) (Fig. 3a for A549, I; b for SPC-A1, II). When the samples were detected for N-cadherin, we found increased levels of N-cadherin in SPC-A1 cell line exposed to PMNs, both in direct (p=0.001 for direct co-culture/monoculture, p= 0.010 for direct/indirect co-culture) and indirect co-culture (p=0.024 for indirect co-culture/monoculture) (Fig. 3d, IV). However, there was no significantly statistical difference of Ncadherin expression in A549 cell lines (p=0.105 for direct coculture/monoculture, p=0.186 for direct/indirect co-culture, p=0.534 for indirect co-culture/monoculture) (Fig. 3c, III). Vimentin was also up-modulated in co-culture of A549/ SPC-A1 with PMNs, especially in direct co-culture (each two group p<0.05) (Fig. 3e for A549, V; f for SPC-A1, VI). Compared to monoculture, cytoplasmic β-catenin was downregulated, and nuclear β-catenin was up-regulated in coculture A549/SPC-A1 cell lines with PMNs, which was indicative of EMT (Fig. 3 IX for A549; X for SPC-A1). Each band of target proteins was integrated and normalized to GAPDH expression, and 11 experiments were performed for each cell line. PMNs influence the migration activity of tumor cells in vitro Accumulating evidences indicated the unfavorable prognostic value of TAN in different types of solid cancers based on the following findings: (i) EMT is an important process of tumor cells migrated to distant sites [9, 14] and (ii) PMNs induce EMT in lung adenocarcinoma cell lines (see above). We hypothesized that PMNs might promote the migration activity of the tumor cells. Under co-culture with PMNs, we detected the migration activities of A549/SPC-A1 cell lines in a timecourse by migration assays. The results showed that in coculture group, the migration activity of tumor cells was strengthened than that of monoculture, particularly in direct co-culture group (all p<0.05 both for A549/SPC-A1) (Fig. 4). There was a difference in the migratory potential of different cell lines so this phenomenon could exist in 8 h for A459 cell lines, while 24 h for SPC-A1. These findings indicated that the migration activity, as an important functional consequence of tumor cells, was stimulated by co-culture with PMNs. TGF-β/Smad signaling pathway Finally, we focused on the possible mechanisms of PMNs promoting EMT in vitro. We initially assessed the levels of TGF-β1 in cell supernatants by ELISA and evaluated whether TGF-β signaling pathway was involved in PMNs promoting EMT. We demonstrated that the concentrations of TGF-β1

Tumor Biol.

Fig. 1 The relationships between haemoptysis and NEUT count were shown in a (p = 0.006 for NEUT), and the y axis represented the absolute value of circulating NEUT (×10 9 /L). A total of 149 haemoptysis and 517 non-haemoptysis patients were enrolled in this analysis. b The Spearman rank correlation test of PMN infiltration and E-cadherin expression. The y axis represented the status of PMN infiltration and E-cadherin expression. The PMN infiltration: 0, negative

infiltration; 1, low infiltration; and 2, high infiltration. The E-cadherin expression: 0, negative expression; 1, weak expression; 2, medium expression; and 3, strong expression. TGF-β1 levels in monoculture and coculture were shown in c and d (c p=0.023 for A549, d p=0.016 for SPCA1), which suggested that the concentration of TGF-β1 was significantly up-modulated in co-culture group, compared to monoculture group

were significantly up-modulated in co-culture groups for both A549 and SPC-A1 cell lines (Fig. 1c: p=0.023 for A549; d: p=0.016 for SPC-A1). Next, we tested whether Smad4, a classical downstream factor of TGF-β signal pathway, was

stimulated under co-culture. We assessed the expressions of Smad4 in cytoplasmic or nuclear localization by WB. We demonstrated that cytoplasmic Smad4 of tumor cells was decreased during co-culture with PMNs. Accordingly, Smad4 in

Fig. 2 The expressions of intratumoral neutrophils (a, c) and E-cadherin (b, d) in lung adenocarcinoma tissue (200×). a Intratumoral neutrophils were absent, whereas the expression of E-cadherin was strong in the same sample (b). c Intratumoral neutrophils were present, whereas the expression of E-cadherin was weak in the same sample (d)

Tumor Biol.

Fig. 3 The results of protein expressions were shown in I–XII, and the corresponding quantified results were shown in a–f. Co-culture resulted in the decrease of E-cadherin, with direct co-culture group declined sharply than indirect group (I, a for A549, II, b for SPC-A1). The expressions of N-cadherin and vimentin in both A549 and SPC-A1 cell lines were increased (except N-cadherin in A549); these changes were more obvious in direct co-culture group than indirect group (A549: III, c for N-cadherin and V, e for vimentin; SPC-A1: IV, d for N-cadherin and VI, f for

vimentin). GAPDH were shown in VII for A549 and VII for SPC-A1. Cytoplasmic β-catenin was down-regulated, and nuclear β-catenin was up-regulated in co-culture A549 (IX) / SPC-A1 (X) cell lines with PMNs. Taken together, PMNs promote EMT in vitro. Upon co-culture, Smad4 in cytoplasm localization was decreased, whereas Smad4 in nuclear localization was increased, which suggested that PMNs might promote EMT via TGF-β/Smad signaling pathway in vitro (XI for A549, XII for SPCA1)

nuclear localization was increased, which suggested that TGF-β/Smad signaling pathway was initiated in the process EMT induced by PMNs in vitro (Fig. 3 XI for cytoplasmic Smad4; XII for nuclear Smad4).

TANs induced EMT of lung cancer cells in vitro and promoted their migration. In addition, TGF-β/Smad signaling pathway was activated. This was the first study focused on the mechanism of haemoptysis as an independent predictor, as far as we know. Circulating neutrophils are drawn across the vasculature towards tumor tissue by chemotactic factors, which are secreted by tumor cells and mesenchymal cells in tumor microenvironment [15, 16]. These chemotactic factors bind to the Gprotein-coupled receptors CXCR1 and CXCR2 expressed on the cell surface of neutrophils [15, 16]. Notably, PMN infiltration was also correlated with the expression of E-cadherin in tissue level. E-cadherin is a well-established indicator of EMT that initiates a series of signaling events and major cytoskeletal reorganization [10]. A correlation between PMN infiltration and E-cadherin leads to the presumption that PMNs could induce EMT by the reciprocity of PMNs and

Discussion This study was a further series about the prognostic role of haemoptysis and built on our previous experience that haemoptysis is an independent prognostic factor in lung adenocarcinoma. In the present study, we explored the mechanisms of haemoptysis associated with dismal outcome and demonstrated that haemoptysis significantly associated with the increased incidence of intratumoral PMN infiltration, which was also related to high baseline NEUT count from lung adenocarcinoma patients. Further study disclosed that

Tumor Biol. Fig. 4 The results of migration assays in A549 (a–c) and SPC-A1 (d–f). The corresponding quantified results were shown in g for A549 and h for SPC-A1. The mean number of migrating tumor cells in three fields was highest in direct co-culture group, while it was lowest in monoculture group

tumor cells in tumor microenvironment. Our presumption was supported by the fact that under co-culture of A549/SPC-A1 cell lines with PMNs from lung cancer patients, the expression of E-cadherin was decreased significantly; whereas Ncadherin and vimentin were increased (There was an exception that the expression of N-cadherin in A549 had no tendency.). And these phenomena in the direct co-culture group were more obvious than those in the indirect group. In the direct co-culture system, PMNs were added to the tumor cells directly, and before further experiments, the non-adherent suspension PMNs were cleared off by washing with PBS carefully. Therefore, the adherent tumor cells were isolated from PMNs. It was inevitable that there were a small number of PMN residues, and these might be reflected in the WB as the PMNs would contribute to the total protein content. But the reflection was weak because there were

only few PMN residues after thorough washing. Another convincing evidence was that a nuclear accumulation of β-catenin was observed in co-culture group. In parallel, the cytoplasm expression of β-catenin declined. These findings collectively indicated that PMNs induced EMT in lung adenocarcinoma cells in vitro. According to the results drawn above, we then asked the question whether PMNs might alter the function consequences of the tumor cells, for instance, the migration activity. Under different culture conditions, the cancer cells that migrated through the transwell were varied in a certain time. The highest count was seen in the direct co-culture group, and the lowest was seen in the monoculture group. Previous studies showed that PMNs enhanced the migration of head and neck cancer (HNC) cell lines through the activation of cortactin, which is an actin-binding protein required for cellular migration and invasion [17].

Tumor Biol.

The reciprocities between cancer cells and PMNs were identified mainly in two ways: direct cell-to-cell contact manner and indirect cytokine-dependent manner. The direct coculture group was established to simulate the cell-tocell contact, and indirect co-culture group simulated the indirect way dependent on secreting cytokines. Taken together, our results demonstrated that both direct and indirect manner contributed to the interactions between tumor cells and PMNs. TGF-β plays important roles in malignant tumor initiation and progression, functioning as both a suppressor and a promoter [18]. It has also been identified to promote tumor growth by inducing EMT [9, 14]. Smad proteins, which are the classical intracellular effectors of TGF-β signaling, are activated by type-II/I TGF-β receptors and into the nucleus to regulate transcription of target genes. In this complicated process, Smad4 was involved in a complex with phosphorylated Smad2/3 and translocated into the nucleus from cytoplasm to regulate transcription of target genes [10]. However, to our knowledge, the role of Smad-dependent TGF-β signaling pathway in the interaction of lung adenocarcinoma cells and PMNs is not clear. In the present study, the production of TGF-β was evoked in co-culture group. Translocation into the nucleus of Smad4 demonstrated that TGF-β/Smad signaling pathway might partly involve in the EMT of lung adenocarcinoma cells promoted by PMNs. Studies have demonstrated that TGF-β has dual functions in formation of pro-tumorigenic (N2) phenotype and antitumorigenic (N1) phenotype [5]. It is also likely that the characters of both N1 and N2 PMNs might be functions of differential neutrophil activation status [1]. Taken together, our findings hypothesized a novel mechanism of tumor-PMN interaction mediated by TGF-β. However, there are a few limitations in our study. Further study to investigate TGF-β/Smad signal pathway by interfering TGF-β1 signal pathway in tumor cells will help to better explain the precise mechanism in this interaction process. In summary, we provided evidence that lung adenocarcinoma with haemoptysis was associated with more PMN infiltration, and co-culture PMNs with tumor cells induce the EMT of tumor cells, TGF-β/Smad signal pathway involved in the process. This may partly provide mechanistic reasons for why haemoptysis was associated with poor outcome in lung adenocarcinoma. Acknowledgments The work was supported by the Shandong Provincial Natural Science Foundation of China (ZR2013HZ001) and National Natural Science Foundation of China (81301728). The authors would like to thank all the members of our team.

Conflict of interest None Financial support The work was supported by the Shandong Provincial Natural Science Foundation of China (ZR2013HZ001) and National Natural Science Foundation of China (81301728).

References 1. 2. 3. 4.

5.

6.

7.

8. 9. 10. 11.

12.

13.

14.

15.

16.

17.

18.

Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–74. Nathan C. Neutrophils and immunity: challenges and opportunities. Nat Rev Immunol. 2006;6:173–82. Donskov F. Immunomonitoring and prognostic relevance of neutrophils in clinical trials. Semin Cancer Biol. 2013;23:200–7. Hu P, Wang G, Cao H, Ma H, Sui P, Du J. Haemoptysis as a prognostic factor in lung adenocarcinoma after curative resection. Br J Cancer. 2013;109:1609–17. Fridlender ZG, Sun J, Kim S, Kapoor V, Cheng G, Ling L, et al. Polarization of tumor-associated neutrophil phenotype by TGF-beta: BN1^ versus BN2^ TAN. Cancer Cell. 2009;16:183–94. Galdiero MR, Garlanda C, Jaillon S, Marone G, Mantovani A. Tumor associated macrophages and neutrophils in tumor progression. J Cell Physiol. 2013;228:1404–12. Grosse-Steffen T, Giese T, Giese N, Longerich T, Schirmacher P, Hansch GM, et al. Epithelial-to-mesenchymal transition in pancreatic ductal adenocarcinoma and pancreatic tumor cell lines: the role of neutrophils and neutrophil-derived elastase. Clin Dev Immunol. 2012;2012:720768. Thiery JP. Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer. 2002;2:442–54. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. J Clin Invest. 2009;119:1420–8. Xu J, Lamouille S, Derynck R. TGF-beta-induced epithelial to mesenchymal transition. Cell Res. 2009;19:156–72. Fuxe J, Karlsson MC. TGF-beta-induced epithelial-mesenchymal transition: a link between cancer and inflammation. Semin Cancer Biol. 2012;22:455–61. Wang LG, Ni Y, Su BH, Mu XR, Shen HC, Du JJ. MicroRNA-34b functions as a tumor suppressor and acts as a nodal point in the feedback loop with Met. Int J Oncol. 2013;42:957–62. Remmele W, Stegner HE. [Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue]. Pathologe. 1987;8:138–40. Do TV, Kubba LA, Du H, Sturgis CD, Woodruff TK. Transforming growth factor-beta1, transforming growth factor-beta2, and transforming growth factor-beta3 enhance ovarian cancer metastatic potential by inducing a Smad3-dependent epithelial-tomesenchymal transition. Mol Cancer Res. 2008;6:695–705. Dumitru CA, Lang S, Brandau S. Modulation of neutrophil granulocytes in the tumor microenvironment: mechanisms and consequences for tumor progression. Semin Cancer Biol. 2013;23:141–8. Tazzyman S, Niaz H, Murdoch C. Neutrophil-mediated tumour angiogenesis: subversion of immune responses to promote tumour growth. Semin Cancer Biol. 2013;23:149–58. Dumitru CA, Bankfalvi A, Gu X, Eberhardt WE, Zeidler R, Lang S, et al. Neutrophils activate tumoral cortactin to enhance progression of orohypopharynx carcinoma. Front Immunol. 2013;4:33. Massague J. TGF-beta in cancer. Cell. 2008;134:215–30.