Inhibition of iNOS as a novel effective targeted therapy against triple-negative breast cancer

Abstract

Introduction: Triple-negative breast cancer (TNBC) is an aggressive form of breast cancer with no effective targeted therapy. Inducible nitric oxide synthase (iNOS) is associated with poor survival in patients with breast cancer by increasing tumor aggressiveness. This work aimed to investigate the potential of iNOS inhibitors as a targeted therapy for TNBC. We hypothesized that inhibition of endogenous iNOS would decrease TNBC aggressiveness by reducing tumor initiation and metastasis through modulation of epithelial-mesenchymal transition (EMT)-inducing factors.

Methods: iNOS protein levels were determined in 83 human TNBC tissues and correlated with clinical outcome. Proliferation, mammosphere-forming efficiency, migration, and EMT transcription factors were assessed in vitro after iNOS inhibition. Endogenous iNOS targeting was evaluated as a potential therapy in TNBC mouse models.

Results: High endogenous iNOS expression was associated with worse prognosis in patients with TNBC by gene expression as well as immunohistochemical analysis. Selective iNOS (1400 W) and pan-NOS (L-NMMA and L-NAME) inhibitors diminished cell proliferation, cancer stem cell self-renewal, and cell migration in vitro, together with inhibition of EMT transcription factors (Snail, Slug, Twist1, and Zeb1). Impairment of hypoxia-inducible factor 1α, endoplasmic reticulum stress (IRE1α/XBP1), and the crosstalk between activating transcription factor 3/activating transcription factor 4 and transforming growth factor β was observed. iNOS inhibition significantly reduced tumor growth, the number of lung metastases, tumor initiation, and self-renewal.

Conclusions: Considering the effectiveness of L-NMMA in decreasing tumor growth and enhancing survival rate in TNBC, we propose a targeted therapeutic clinical trial by re-purposing the pan-NOS inhibitor L-NMMA, which has been extensively investigated for cardiogenic shock as an anti-cancer therapeutic.

Figures

Figure 1

Enhanced nitric oxide synthase 2 (NOS2) expression correlates with poor patient survival in invasive triple-negative breast cancer (TNBC). (A) Higher NOS2 mRNA expression in invasive TNBC versus non-TNBC (P = 3.85 × 10−5). (B) High NOS2 expression correlates with death at 5 years in invasive breast carcinoma (P = 0.037). Kaplan-Meier survival analyses in (C) Van de Vijver (n = 69; P = 0.04) and (D) Curtis (n = 260; P = 0.01) breast databases show that high NOS2 expression correlates with worse overall survival of patients with TNBC. (E) Immunohistochemical analysis of TNBC human samples for inducible nitric oxide synthase (iNOS) protein expression. Weak-to-moderate (3 or 4), moderate-to-strong (5 or 6), and strong (7) were the cutoffs established for further analysis of survival. Several samples showed expression in both tumor (T) and stromal (S) cells (original optical objective: 20×). MDA-MB-231 cells transfected with either NOS2-directed shRNA (shRNA1) or empty vector (EV) were used as negative and positive control of iNOS staining, respectively (original optical objective: 10×). Counterstain: hematoxylin. (F) Increased iNOS expression is associated with less patient survival when compared with low iNOS expression. Kaplan-Meier survival analysis of TNBC human patient samples (n = 83) (P = 0.05). shRNA1, small hairpin RNA 1; TCGA, The Cancer Genome Atlas.

Figure 2

Effects of inducible nitric oxide synthase (iNOS) inhibitors on tumor cell proliferation, migration, and mammosphere formation in triple-negative breast cancer (TNBC) cell lines. Proliferation (A, B) and primary (C) and (D) secondary mammosphere, and migration index of MDA-MB-231 and SUM159 cell lines treated with 1400 W (E) and L-NMMA (F) for 96 hours. Results were normalized to vehicle. Data are presented as mean ± standard error of the mean. *P <0.05, **P <0.01, ***P <0.001, ****P <0.0001. 1400 W, N-[[3-(aminomethyl)phenyl]methyl]-ethanimidamide; L-NMMA, NG-monomethyl-L-arginine; MSFE, mammosphere-forming efficiency.

Figure 3

iNOS knockdown impairs tumorigenicity and EMT by a dual impact on HIF1α and ER stress/TGFβ/AFT4/ATF3 crosstalk. Proliferation (A), migration (B), and mammosphere-forming efficiency (MSFE) (C) in MDA-MB-231 cells transiently transfected with two different NOS2-directed siRNAs (siRNA1 and siRNA2) compared with scrambled control. Western blot analysis of NOS isoforms (iNOS, eNOS, and nNOS) and EMT transcription factors in MDA-MB-231 and SUM159 cell lines treated with (D) 1400 W and (E) siRNA-mediated NOS2 knockdown. (F) Selective iNOS inhibition reduced hypoxia (HIF1α) and ER stress markers (IRE1α and ATF4). (G) Phospho-Smad2/3, Smad2/3, and mature TGFβ protein levels in MDA-MB-231 and SUM159 cells. (H) Recombinant TGFβ1 (10 ng/mL for 7 days) activates the PERK/eIF2α/ATF4/ATF3 axis in MCF10A. (I) Effects on the PERK/eIF2α/ATF4/ATF3 axis by co-treatment of recombinant TGFβ1 (10 ng/mL, 7 days) and 1400 W (4 mM) for 24 hours in MCF10A cells. iNOS, ATF4, ATF3, and mature TGFβ protein levels in siRNA-mediated NOS2 knockdown (siRNA2) MCF10A cells for 96 hours. (J) Selective iNOS inhibition is postulated to impair EMT and tumor cell migration by an impact on HIF1α, ER stress (IRE1α/XBP1), and the crosstalk between ATF4, ATF3, and TGFβ. Results were normalized to scrambled. Data are presented as mean ± standard error of the mean. ****P <0.0001. 1400 W, N-[[3-(aminomethyl)phenyl]methyl]-ethanimidamide; ATF3, activating transcription factor 3; ATF4, activating transcription factor 4; EMT, epithelial-mesenchymal transition; ER, endoplasmic reticulum; HIF1α, hypoxia-inducible factor 1α; iNOS, inducible nitric oxide synthase; siRNA, small interfering RNA; TGFβ, transforming growth factor β.

Figure 4

Decrease in tumor initiation and lung metastases in MDA-MB-231 xenografts. (A) Tumor volume of MDA-MB-231 breast xenografts (n = 5 per group) after daily injection of L-NAME (80 mg/kg, intraperitoneal). (B) Primary and secondary mammosphere of cancer cells isolated from tumor tissue. (C) Luminescence of MDA-MB-231 L/G tumor cells in lungs of vehicle- and L-NAME-treated mice. (D) Tumor-initiating capacity of tumor cells (limiting dilution assay). Results were normalized to vehicle. Data are presented as mean ± standard error of the mean. *P <0.05, **P <0.01, ***P <0.001. L-NAME, N5-[imino(nitroamino)methyl]-L-ornithine methyl ester; MSFE, mammosphere-forming efficiency.

Figure 5

In vivoeffects of L-NMMA in MDA-MB-231 xenografts. (A) Tumor volume of MDA-MB-231 breast xenografts (n = 10 per group) treated with vehicle, L-NMMA, docetaxel, and combination. (B) Illustrative images of Ki67 staining in the vehicle, L-NMMA, docetaxel, and combination groups. Original optical objective: 10×. Counterstain: hematoxylin. (C) Cell proliferation of tumor xenografts with Ki67 immunostaining. (D) Nuclear cleaved caspase-3 staining in the chemo and combo groups. (E) Primary and secondary mammosphere of breast cancer cells isolated from tumor tissue. (F) Tumor-initiating capacity of tumor cells. Results were normalized to vehicle. For proliferation and apoptosis, 1,000 cells were counted from 10 different fields, and the percentage was determined. Data are presented as mean ± standard error of the mean. *P <0.05, **P <0.01, ***P <0.001, ****P <0.0001. L-NMMA, NG-monomethyl-L-arginine; MSFE, mammosphere-forming efficiency.

Figure 6

Clinically relevant dose regimen of L-NMMA in orthotopic mouse models of triple-negative breast cancer. (A) Mean systolic pressure of mice (n = 5) giving one cycle of the dose rate proposed in this study. (B) Mean systolic pressure of mice 30 minutes and 24 hours after the last injection of one cycle treatment (n = 5). (C) Tumor volume of MDA-MB-231 xenografts (n = 10 per group) treated with vehicle, docetaxel, and combination (docetaxel + amlodipine + L-NMMA). (D) Kaplan-Meier survival curve of vehicle-, chemotherapy-, and combo-treated MDA-MB-231 xenograft-bearing mice. (E) Tumor volume of SUM159 xenografts (n = 10 per group) treated with vehicle, docetaxel, and combination. (F) Effects of amlodipine on tumor volume in MDA-MB-231 xenografts (n = 5 per group). Data are presented as mean ± standard error of the mean. *P <0.05, **P <0.01, ***P <0.001, ****P <0.0001. L-NMMA, NG-monomethyl-L-arginine; ns, not significant.