Activation of EGFR, HER2 and HER3 by neurotensin/neurotensin receptor 1 renders breast tumors aggressive yet highly responsive to lapatinib and metformin in mice

Abstract

A present challenge in breast oncology research is to identify therapeutical targets which could impact tumor progression. Neurotensin (NTS) and its high affinity receptor (NTSR1) are up regulated in 20% of breast cancers, and NTSR1 overexpression was shown to predict a poor prognosis for 5 year overall survival in invasive breast carcinomas. Interactions between NTS and NTSR1 induce pro-oncogenic biological effects associated with neoplastic processes and tumor progression. Here, we depict the cellular mechanisms activated by NTS, and contributing to breast cancer cell aggressiveness. We show that neurotensin (NTS) and its high affinity receptor (NTSR1) contribute to the enhancement of experimental tumor growth and metastasis emergence in an experimental mice model. This effect ensued following EGFR, HER2, and HER3 over-expression and autocrine activation and was associated with an increase of metalloproteinase MMP9, HB-EGF and Neuregulin 2 in the culture media. EGFR over expression ensued in a more intense response to EGF on cellular migration and invasion. Accordingly, lapatinib, an EGFR/HER2 tyrosine kinase inhibitor, as well as metformin, reduced the tumor growth of cells overexpressing NTS and NTSR1. All cellular effects, such as adherence, migration, invasion, altered by NTS/NTSR1 were abolished by a specific NTSR1 antagonist. A strong statistical correlation between NTS-NTSR1-and HER3 (p< 0.0001) as well as NTS-NTSR1-and HER3- HER2 (p< 0.001) expression was found in human breast tumors. Expression of NTS/NTSR1 on breast tumoral cells creates a cellular context associated with cancer aggressiveness by enhancing epidermal growth factor receptor activity. We propose the use of labeled NTS/NTSR1 complexes to enlarge the population eligible for therapy targeting HERs tyrosine kinase inhibitor or HER2 overexpression.

Figure 1. NTS/NTSR1 complex enhanced experimental tumor growth generated in human breast cancer cell lines

(A) Experimental tumors were generated from the breast cancer cell line, MCF-7 and the NTS-overexpressing subclones. Comparative growth curves of MCF-7, NTS-h (high NTS expression) and NTS-l (low NTS expression) cells xenografted in 10, 20, and 25 mice, respectively. Tumor volumes were measured every week. Inset, NTS and NTSR1 transcript analysis from 200ng of MCF-7, NTS-h, and NTS-l total RNA. (B) Proportion of animals with metastases, and metastases distribution within organs and lymph nodes. (C) Typical H&E staining performed on paraffin sections of (left) invaded lymph node at 50X magnification, 200X magnification (black square) or 400X magnification (white square); (middle) lung metastasis at 400X magnification; (right) kidney metastasis at 400X magnification.

Figure 2. NTS autocrine and paracrine regulation enhanced EGFR, HER2, and HER3 basal expression and activation in human breast cancer cell lines

(A) HER2 and HER3 immunohistochemistry performed on paraffin embedded tumors from mice xenograph with MCF-7, NTS-l or NTS-h. 200X magnification and computer enlargement of specific areas. (B) Breast cancer cells MCF-7, NTS-l and NTS-h, with the histograms representing intensity-based quantification of Western blot bands of basal total protein, EGFR, HER2, and HER3, using Morpho Expert software (Explora Nova, France). Values are expressed as the percentage of the control MCF-7 cells and are the mean ± SEM of 5 to 7 independent experiments. (C) Representative western blot analyses of EGFR, HER2, HER3 and ERK 1/2 total protein from MCF-7 and NTS-l cells treated with 5x10⁻⁶ M SR 48692. (D) EGFR, HER2, and HER3 immunolabeling in MCF-7 and NTS-l cells treated after 48h of seeding. (E) Breast cancer cells MCF-7 and NTS-l, with the histograms representing intensity-based quantification of Western blot bands of phosphorylated protein, EGFR, HER2, and HER3. Values are expressed as the percentage of the control MCF-7 cells and are the mean ± SEM of 5 to 7 independent experiments. (F) Representative western blot analyses of Phospho EGFR, phosphoHER2, and Phospho HER3 protein from MCF-7 and NTS-l cells treated with DMSO or 5x10⁻⁶ M SR 48692 for 48h.

Figure 3. EGF like ligands and metalloprotease released by NTS

(A) Amount of Hb-EGF (pg/ml), assayed in 0% FCS culture media of MCF-7, or NTS- l, cells. Cells were not treated, or treated for 24h with DMSO, 5x10-6M, SR 48692. Using Paired t test p = 0.0088 between DMSO and SR 48692 NTS-l treated cells, n=4; with unpaired test p=0.03 between MCF-7 and NTS-l, n=5. (B) Amount of NGR2 (pg/ml) assay in 0% FCS culture media of MCF-7 or NTS-l cells not treated or treated for 48 h with DMSO, 5x10-6M SR 48692. Using Paired t test p = 0.005 between DMSO and SR 48692 NTS-l treated cells, n=5; with unpaired test p=0.016 between MCF-7 and NTS-l, n=4. (C) MMP9 transcript analysis of total RNA from MCF-7, and NTS-l treated with DMSO or 5x10⁻⁶ M SR 48692 for 48h. (D) Amount of MMP9 (pg/ml) assay in 0% FCS culture media of MCF-7 or NTS-l cells not treated or treated for 48 h with DMSO, 5x10⁻⁶M SR 48692. Using Paired t test p = 0.005 between DMSO and SR 48692 NTS-l treated cells, n=5; with unpaired test p=0.003 between MCF-7 and NTS-l, n=5.

Figure 4. NTS/NTSR1 expressing tumors response to EGFR/HER2 inhibitor treatments

(A) NTS-h cells were inoculated in the left mammary gland of the mice. Here is shown an example of a mouse from each group after 23 days of treatment. (B) Tumor growths generated by NTS-h cells treated for 23 days with sesame oil 6% DMSO, or 75 mg/kg lapatinib, or 200 mg/kg metformin, or both. At day one, 7 mice per group were randomized on tumors size reaching approximately 95 mm³.

Figure 5. NTS autocrine and paracrine regulations enhanced oncogenic cellular effects on EGF-induced migration and invasion

(A) Number of colonies formed on semi-solid medium after 12 days expressed as the percentage of MCF-7 cells. (B) Adhesion assays were performed on type I collagen supports. After 1h or 48h of seeding, cells were gently washed and the remaining attached cells were quantified by spectrophotometric analysis of crystal violet staining. Results represent the mean optic density ± SEM of 4 experiments. (C) Speed of migration on type I collagen of MCF-7 and NTS-l cells, control or treated with EGF (10 ng/mL), in the presence or not of SR 48692 (5x10-6 M). Results represent the mean ± SEM of 9 to 10 independent experiments. (D) Speed of migration on type I collagen of MCF-7 and NTS-l cells control or treated with EGF (10 ng/mL), in the presence or not of the PKC-inhibitor Gö6976 (5x10-8M) or the PLC-inhibitor U73122 (5x10-6 M) Results represent the mean ± SEM of 4 independent experiments. Results represent the mean ± SEM of 3 to 4 experiments. Student-Newman-Keuls Multiple Comparisons Test was performed on the data : ***P<0.001,**P<0.01, and *P<0.05. (E) Synergism between NTS and EGF on invasion in a type I collagen invasion assay. Cells were seeded on the top of a type I collagen gel and treated with EGF (100 ng/mL) in the presence or absence of Gö6976 (5x10-8 M). Results represent the mean ± SEM of 3 to 4 experiments.

Figure 6. Synergy between NTS and EGF to activate EGFR, HER2, and HER3

(A) Breast cancer cells NTS-l or MCF-7, with the histograms representing intensity- based quantification of Western blot bands of phosphorylated protein, EGFR, HER2, and HER3 treated for 10 min with 10ng/ml EGF. Values are expressed as the percentage of the EGF treated MCF-7 cells and are the mean ± SEM of 5 independent experiments. (B) Representative Western blot analyses of phosphoEGFR, phosphoHER2, phosphoHER3 and actin from MCF-7 and NTS-l cells treated or not with 10ng/ml EFG for 10 min.

Figure 7. NTS, NTSR1, HER2, and HER3 immunohistochemistry on breast and lung cancer tumors

Two examples of positive labeling scored 2 or 3 for NTS, NTSR1, HER2, HER3 from breast tumor TMA, labeling was performed on consecutive slides. 100X magnification.