Nicotine promotes breast cancer metastasis by stimulating N2 neutrophils and generating pre-metastatic niche in lung

Abstract

Smoking has a profound impact on tumor immunity, and nicotine, which is the major addictive component of smoke, is known to promote tumor progression despite being a non-carcinogen. In this study, we demonstrate that chronic exposure of nicotine plays a critical role in the formation of pre-metastatic niche within the lungs by recruiting pro-tumor N2-neutrophils. This pre-metastatic niche promotes the release of STAT3-activated lipocalin 2 (LCN2), a secretory glycoprotein from the N2-neutrophils, and induces mesenchymal-epithelial transition of tumor cells thereby facilitating colonization and metastatic outgrowth. Elevated levels of serum and urine LCN2 is elevated in early-stage breast cancer patients and cancer-free females with smoking history, suggesting that LCN2 serve as a promising prognostic biomarker for predicting increased risk of metastatic disease in female smoker(s). Moreover, natural compound, salidroside effectively abrogates nicotine-induced neutrophil polarization and consequently reduced lung metastasis of hormone receptor-negative breast cancer cells. Our findings suggest a pro-metastatic role of nicotine-induced N2-neutrophils for cancer cell colonization in the lungs and illuminate the therapeutic use of salidroside to enhance the anti-tumor activity of neutrophils in breast cancer patients.

Fig. 1. Nicotine-induced neutrophils facilitate premetastatic niche formation in the lung.

A Lung-metastasis incidence in breast cancer patients cohort (n = 1077) comprising both never and current or ex-smokers (two-sided Fisher exact test with relative risk 1.35 and CI 1.06/1.72). B Upper panel: Schematic diagram for spontaneous metastasis. Mouse mammary carcinoma 4T1 cells (20⁴) were orthotopically implanted into mice MFP followed by i.p. injection of PBS (n = 7) or nicotine (n = 6) every other day till the endpoint. Lower panel: Representative images of in vivo primary tumor (top) and ex vivo metastatic lungs (bottom). C Ex vivo quantification of lung metastasis in PBS (n = 7) or nicotine-treated (n = 5) mice by bioluminescence imaging (BLI) at the endpoint (unpaired two-tailed t-test, *p = 0.01). D In vivo growth of primary tumor in PBS (n = 7) or nicotine-treated (n = 6) mice as quantified by BLI (two-way ANOVA and Tukey’s multiple comparisons test, *p = 0.04, ***p = 0.0001). E Primary tumor weight in PBS (n = 7) or nicotine-treated (n = 5) mice (unpaired two-tailed t-test, *p = 0.02). F Kaplan–Meier analysis of lung metastasis-free survival in PBS (n = 7) or nicotine-treated (n = 6) mice [log-rank (Mantel–Cox test), **p = 0.001]. G Quantification of IHC staining [number of cell infiltrated/high power field (HPF)] in primary tumor and metastatic lungs in PBS or nicotine-treated mice (n = 5 mice/group, unpaired two-tailed t-test, ***p = 0.0002, ****p < 0.0001, *p = 0.04, p = 0.57). H Upper panel: Schematic diagram for premetastatic treatment with nicotine. Balb/c mice were intraperitoneally injected with PBS or nicotine (2 mg/kg, n = 8/group) every other day till day 10, followed by i.v. injection of 4T1 cells (10⁴) into tail vein. Lower panel: Representative images of in vivo (top) and ex vivo metastatic lungs (bottom). I, J In vivo and ex vivo quantification of lung metastasis incidence in PBS or nicotine-pretreated mice by BLI [n = 8 mice/group, two-way ANOVA and Tukey’s multiple comparisons test (in vivo), *p = 0.04, ***p = 0.0001, and Wilcoxon rank-sum test (ex vivo), *p = 0.01]. K Representative H&E and IHC images of neutrophils (ELA2) in PBS (control) or nicotine-pretreated premetastatic (w/o tumor cell) and metastatic lungs (w/ tumor cells) (n = 3/individual experiment; scale bar: 100 µm). All data are presented as mean ± S.E.M. ns, not significant; MFP, mammary fat pad; i.p., intraperitoneal; i.v., intravenous.

Fig. 2. Blocking nicotine-induced neutrophils infiltration decreases lung metastasis.

A Upper panel: Schematic diagram of neutrophil depletion experiment. 4T1 cells (10⁴) were intravenously injected into the tail vein of Balb/c mice as indicated with or without i.p. injection of nicotine (2 mg/kg) every other day plus either isotype or anti-Ly6G antibody injection every 3 days. Lower panel: Representative images of in vivo (top) and ex vivo lung metastasis (bottom). B In vivo quantification of lung metastasis incidence with or without nicotine plus isotype or anti-Ly6G treated Balb/c mice by BLI (n = 14 for isotype and n = 15 for anti-Ly6G, two-way ANOVA and Tukey’s multiple comparisons test, ****p < 0.0001, p = 0.45, *p = 0.01). C Ex vivo quantification of lung metastasis with or without nicotine plus isotype or anti-Ly6G treated Balb/c mice by BLI at the endpoint (n = 14 for isotype and n = 15 for anti-Ly6G, unpaired two-tailed t-test, **p = 0.006, **p = 0.003). D Flow cytometric quantification of lung infiltrating CD11b⁺ cells (neutrophils) in control, tumor-bearing nicotine-free, nicotine-treated tumor-free, and tumor-bearing mice treated with isotype or anti-Ly6G (n = 3 mice/group, unpaired two-tailed t-test, ***p = 0.0001, **p = 0.001, p = 0.10, **p = 0.002, ***p = 0.0006, p = 0.68). E Flow cytometric quantification of total neutrophils (CD11b⁺Ly6G⁺) and monocyte (CD11b⁺Ly6C⁺) in control (tumor and nicotine-free), tumor-bearing nicotine-free, nicotine-treated tumor-free, and tumor-bearing mice treated with isotype or anti-Ly6G antibody (n = 3 mice/group, unpaired two-tailed t-test, ****p < 0.0001, *p = 0.01, **p = 0.007, **p = 0.003, ***p = 0.0001, ***p = 0.0002, p = 0.12). F Average number of lung metastases nodules with or without nicotine plus isotype or anti-Ly6G treated Balb/c mice (n = 14 for isotype and n = 15 for anti-Ly6G, unpaired two-tailed t-test, *p = 0.01, ****p < 0.0001). G Kaplan–Meier analysis of lung metastasis-free survival with or without nicotine plus isotype or anti-Ly6G-treated mice [log-rank (Mantel–Cox test), *p = 0.01, ***p = 0.0002]. All data are presented as mean ± S.E.M. ns, not significant; i.p., intraperitoneal; i.v., intravenous.

Fig. 3. Nicotine promotes N2-neutrophil polarization via STAT3 activation.

A, B Human primary neutrophils and HL-60 were treated with or without nicotine (1 µM) for 12 h. RNA were prepared and examined for the expression of N1 (CD95, NOS2, CCL3), N2 (CD206, CCL2, ARG2) markers and for STAT3 expression by qRT-PCR. β-Actin was used as a normalization control [n = 3 individual experiment/group, unpaired two-tailed t-test, **p = 0.001, *p = 0.01, *p = 0.03, **p = 0.005, ***p = 0.0003 (primary), *p = 0.01, ***p = 0.0002 (HL-60)]. C HL-60 neutrophils were treated with nicotine (1 µM) or STATTIC (1 µM) for 24 h, and examined for p-STAT3 expression by western blot. GAPDH was used as controls (n = 3 individual experiment/group). D Left panel: Schematic diagram for adoptive transfer (ADT) of nicotine-polarized N2-enriched neutrophils. 4T1 cells (10⁴) were intravenously injected into tail vein of athymic nude mice followed by i.p. injection of nicotine (2 mg/kg, n = 6) every other day till day 9 or nicotine plus STATTIC (10 mg/kg, n = 13) every other day from day 9 to 18. At day 20, adoptive transfer was performed with N2-enriched neutrophils (10⁷ cells) derived from bone marrow of control or nicotine-treated Balb/c mice as described in “Methods”. Right panel: Representative images of in vivo (top) and ex vivo metastatic lungs (bottom). E, F In vivo and ex vivo quantification of lung metastasis incidence in nicotine (n = 6) or nicotine plus STATTIC (n = 7) or nicotine plus STATTIC plus adoptively transferred N2-neutrophils (n = 6) [two-way ANOVA and Tukey’s multiple comparisons test (in vivo), *p = 0.04, **p = 0.01, p = 0.67, ***p = 0.001, p < 0.0001, and unpaired two-tailed t-test (ex vivo), *p = 0.01, *p = 0.02]. G Average number of lung metastases nodules (n = 6 for nicotine, n = 7 for nicotine plus STATTIC and n = 6 for nicotine plus STATTIC plus N2-ADT, unpaired two-tailed t-test, ****p < 0.0001, *p = 0.04). H Representative H&E (left panel) and immunofluorescence images (right panel) of N1 (NOS2⁺) and N2 (CD206⁺) neutrophils in metastatic lung sections of nicotine or nicotine plus STATTIC or nicotine plus STATTIC plus adoptively transferred N2-neutrophils (n = 3/individual experiment; scale bar: 100 µm). I Quantification of immunofluorescence staining for N1 (NOS2⁺) and N2 (CD206⁺) neutrophils for the panel (H) (n = 3 mice/group, unpaired two-tailed t-test, **p = 0.005, ***p = 0.0008, ****p < 0.0001). All data are presented as mean ± S.E.M. ns, not significant; i.p., intraperitoneal; i.v., intravenous.

Fig. 4. Nicotine-activated N2 neutrophils promote MET phenotype.

A Upper panel: Experimental setup for obtaining control or nicotine-polarized neutrophil conditioned medium. Lower panels: Morphological changes in MDAMB231 and MCF10CA1a cells treated with the indicated CM. B MDAMB231 and MCF10CA1a cancer cells were treated with the indicated nicotine-activated neutrophil CM for 24 h. Cells were then examined for the expression of epithelial (E-cadherin, EpCam, Krt18) and mesenchymal (Vimentin, N-cadherin, ZEB1) markers by qRT-PCR and western blot. β-Actin was used as input control for qRT-PCR. Representative immunoblots shows expression of E-cadherin, Vimentin, and ZEB1 in MDAMB231 and MCF10CA1a. GAPDH was used as a control for western blot [n = 3 individual experiment/group, unpaired two-tailed t-test, ***p = 0.0003, *p = 0.02, *p = 0.01, **p = 0.007 (MDAMB231), ***p = 0.001, ***p = 0.0008, **p = 0.001, ***p = 0.0004 (MCF10CA1a)]. C Representative IHC images for E-cadherin in metastatic lungs derived from the experiment in Figs. 1B and 2A (n = 3/individual experiment; scale bar: 100 µm). D MDAMB231, MCF10CA1a, and MCF7 cancer cells were treated with or without nicotine (1 µM) for 24 h followed by examining the expression of epithelial (E-cadherin, EpCam) and mesenchymal (Vimentin, ZEB1) markers by qRT-PCR. β-Actin was used as a control for normalization [n = 3 individual experiment/group, unpaired two-tailed t-test, *p = 0.03, ****p < 0.0001, **p = 0.003 (MDAMB231), ***p = 0.0003, **p = 0.001, **p = 0.007, *p = 0.03 (MCF10CA1a), **p = 0.004, ***p = 0.0002, **p = 0.001 (MCF7)]. E MDAMB231 cancer cells were treated with either nicotine or with conditioned medium derived from nicotine-treated human primary neutrophils. Cells were then examined for their colony-forming ability (n = 3 individual experiment/group, unpaired two-tailed t-test, **p = 0.003). F Expression of α4 and β2 nicotine receptors in primary neutrophils (human and mouse) and in various breast cancer cells (MCF7, MDAMB231, MCF10CA1a, BT549) was examined by qRT-PCR. β-Actin was used as a control for normalization (n = 3 individual experiment/group). All data are presented as mean ± S.E.M.

Fig. 5. Nicotine-induced N2 neutrophil increases secretion of LCN2 and regulates epithelial plasticity of cancer cells.

A LCN2 protein level in CM derived from nicotine-treated (1 µM) or untreated neutrophils (human primary and HL-60 cells) and breast cancer cells (MDAMB231) was quantified by ELISA (n = 3 individual experiment/group, one-way ANOVA ***p = 0.0003, **p = 0.03, p = 0.08). B MDAMB231 and MCF10CA1 cells were treated with 2.2 µm human recombinant LCN2 and the expression of E-cadherin and Vimentin was examined by qRT-PCR and western blot. β-Actin and GAPDH were used as controls [n = 3 individual experiment/group, unpaired two-tailed t-test, **p = 0.009, **p = 0.001 (MDAMB231), **p = 0.002, ***p = 0.0006 (MCF10CA1a)]. C MDAMB231 cells were treated with or without recombinant LCN2 (2.2 µM) for 24 h, followed by assessing E-cadherin and Vimentin expression by immunocytochemistry (n = 3 individual experiment/group; scale bar: 50 µM). D MDAMB231 cells were treated with indicated CM or nicotine alone or with human recombinant LCN2 (2.2 µM) to rescue from suppressive effect of nicotine after 24 h. Cells were then examined for the expression of E-cadherin, Vimentin, pERK1/2, ERK2, and KLF4 by western blot. GAPDH was used as a control (n = 3 individual experiment/group). E, F LCN2 was quantified in serum and urine samples of stage-II, hormone receptor-negative breast cancer patients (n = 20 for serum and n = 20 for urine) and cancer-free females subjects (n = 12 for serum and n = 20 for urine) with or without smoking history by ELISA [unpaired two-tailed t-test, ****p < 0.0001 (serum), **p = 0.006, ***p = 0.0007 (urine)]. All data are presented as mean ± S.E.M.

Fig. 6. Salidroside selectively suppresses nicotine-induced N2-neutrophil polarization and lung metastasis.

A Experimental diagram for screening of natural compound library containing 124 compounds. B Top 6 natural compounds that showed more than 1000 fold reduction of ARG2 expression by duplex qRT-PCR. C ARG2 protein in CM derived from HL-60 neutrophils was examined by ELISA after treating with or without salidroside (0.1 µM) in the presence or absence of nicotine (1 µM) (n = 3 individual experiment/group, unpaired two-tailed t-test, *p = 0.01, **p = 0.001, *p = 0.04). D Nicotine and nicotine plus salidroside treated cells were examined for expressions of N1 (CD95, NOS2) and N2 (CD206, ARG2) markers by qRT-PCR. β-Actin was used as a control for normalization (n = 3 individual experiment/group, unpaired two-tailed t-test, ***p = 0.0001, ****p < 0.0001). E Cells in (C) were examined for expressions of STAT3 and p-STAT3 by qRT-PCR and western blot. β-Actin and GAPDH were used as controls (n = 3 individual experiment/group, unpaired two-tailed t-test, ***p = 0.0009, *p = 0.03, ***p = 0.0002). F Upper panel: Schematic diagram for experimental metastasis. MCF10CA1a cells (50⁴) were intravenously injected into tail vein of athymic nude mice followed by i.p. injection of nicotine (2 mg/kg, n = 7) or nicotine plus salidroside (20 mg/kg, n = 7) every other day till the endpoint. Lower panel: Representative images of in vivo (top) and ex vivo metastatic lungs (bottom). G–J In vivo quantification of lung metastasis (n = 7 individual mice/group, two-way ANOVA and Tukey’s multiple comparisons test, *p = 0.03, ****p < 0.0001) (G), average number of lung metastases nodules (n = 7 individual mice/group; data are presented as mean ± S.E.M., *p = 0.01, unpaired two-tailed t-test) (H), ex vivo quantification of lung metastasis (n = 7 individual mice in each group, unpaired two-tailed t-test, ***p = 0.0006) (I), and lung metastasis-free survival in nicotine or nicotine plus salidroside treated nude mice [log-rank (Mantel–Cox test), ***p = 0.0003] (J). All data are presented as mean ± S.E.M. i.p., intraperitoneal; i.v., intravenous.

Fig. 7. Proposed model illustrating a nicotine-induced lung metastasis.

Nicotine induces N2-neutrophil polarization in pre-metastatic niche which promotes ERK/KLF4–dependent MET via secreting STAT3-activated LCN2 for cancer cell colonization in the lung.