β2-Adrenoceptors on tumor cells play a critical role in stress-enhanced metastasis in a mouse model of breast cancer

Abstract

Chronic stress accelerates metastasis - the main cause of death in cancer patients - through the activation of β-adrenoceptors (βARs). We have previously shown that β2AR signaling in MDA-MB-231(HM) breast cancer cells, facilitates invadopodia formation and invasion in vitro. However, in the tumor microenvironment where many stromal cells also express βAR, the role of β2AR signaling in tumor cells in metastasis is unclear. Therefore, to investigate the contribution of β2AR signaling in tumor cells to metastasis in vivo, we used RNA interference to generate MDA-MB-231(HM) breast cancer cells that are deficient in β2AR. β2AR knockdown in tumor cells reduced the proportion of cells with a mesenchymal-like morphology and, as expected, reduced tumor cell invasion in vitro. Conversely, overexpression of β2AR in low metastatic MCF-7 breast cancer cells induced an invasive phenotype. Importantly, we found that knockdown of β2AR in tumor cells significantly reduced the impact of stress on metastasis in vivo. These findings highlight a crucial role for β2AR tumor cell signaling in the adverse effects of stress on metastasis, and indicate that it may be necessary to block β2AR on tumor cells to fully control metastatic progression.

Keywords: Breast cancer; Chronic stress; Invasion; Metastasis; β(2)-Adrenoceptor.

Fig. 1. Generation of MDA-MB-231^HM breast cancer cells deficient in β₂AR

(a) Quantification of ADRB mRNA transcript levels using qRT-PCR in MDA-MB-231^HM cells transduced with shRNA scramble sequence (control) or shRNA targeting ADRB2 (shADRB2a and shADRB2b) (n = 4). ND: not detected. (b) Quantification of cAMP accumulation in scramble control, shADRB2a and shADRB2b tumor cells in response to increasing concentrations of the β₂AR selective-agonist formoterol (n = 3). (c) Proliferation of control, shADRB2a and shADRB2b tumor cells (n = 3). Data represent mean ± standard error. **p < 0.01 by one-way ANOVA with post hoc Tukey’s planned comparison tests.

Fig. 2. Characterization of tumor cell invasiveness in β₂AR-deficient MDA-MB-231^HM breast cancer cells

(a) Quantification of MMP2 mRNA transcript levels using qRT-PCR in scramble control, shADRB2a and shADRB2b tumor cells treated with vehicle, isoproterenol (non-selective βAR agonist, ISO, 1 µM) and/or ICI-118,551 (β₂AR-selective antagonist, 1 µM) (n = 3 – 5). (b) Fold change in proportion of tumor cells that exhibit mesenchymal-like morphology (defined by cell roundness < 0.5) in response to ISO (1 µM, normalized to vehicle) (n = 5 – 6). (c) Representative images of mesenchymal-like cells (roundness < 0.5) highlighted in orange, in cultures of scramble control, shADRB2a, shADRB2b cells after treatment with vehicle or 1 µM ISO. Inset: a magnified view of a mesenchymal-like tumor cell. Scale bar: 100 µm). (d) Quantification of invasion through transwells by scramble control, shADRB2a and shADRB2b tumor cells in response to ISO (1 µM) treatment (n = 4). Data represent mean ± standard error. *p < 0.05, **p < 0.01 and ***p < 0.001 by (b) one-way ANOVA with repeated measures or (a and d) two-way ANOVA with post hoc Tukey’s planned comparison tests. # p < 0.05, ### p < 0.001 for ISO vs. ISO + ICI-118551 within each cell line.

Fig. 3. Overexpression of β₂AR in MCF-7 cells increases tumor cell invasiveness

(a and b) Quantification of ADRB mRNA transcript levels using qRT-PCR in (a) MDA-MB-231^HM and MCF-7 wildtype (MCF7-WT) cells (n = 3) and (b) MCF7-WT and MCF-7 overexpressing β₂AR (MCF7-β₂AR) cells (n = 3). Normalized to ACTB levels. (c) Representative images of invadopodia formation assay. Panel i. A cell showing β2AR-GFP (green). Panels ii–iv. A cell showing localization of actin foci (green, ii), matrix degradation (grey patches of degradation on red-fluorescent gelatin, iii), and merged image with inset panel showing active invadopodia that have degraded matrix (iv) Scale bar: 10 µm. (d) Quantification of invadopodia per cell in MCF7-WT cells and MC7-β₂AR cells treated with vehicle, isoproterenol (non-selective βAR agonist, ISO, 0.5 µM) or formoterol (β₂AR-selective agonist, FORM, 0.5 µM) (n > 80 cells per condition). Data represent mean ± standard error. *p < 0.05 and ***p < 0.001 by (a and b) unpaired t-tests or (d) one-way ANOVA with post hoc Tukey’s planned comparison tests.

Fig. 4. β₂AR knockdown in MDA-MB-231^HM breast cancer cells impairs metastasis in vivo

(a) Left panel: Representative in vivo bioluminescence image of the orthotopic MDAMB-231 breast cancer model showing primary tumor (PT) and spontaneously disseminated metastasis in lymph node (LN) and lung 28 days after tumor cell injection. Right panel: Quantification of distant metastasis by bioluminescence imaging in mice with tumors derived from scramble control or β₂AR-deficient cells (shADRB2a and shADRB2b). Mice were exposed to non-stress or stress conditions. (n = 5 at each time point). (b) Representative images of metastasis in vivo at day 14 and 28 after tumor cell injection. Note: the primary tumor is not visible. Scale: min. 3 × 10⁶ photons/sec; max. 1 × 10⁸ photons/sec. (c) The magnitude of the effect of stress on metastatic burden was computed as the difference in the rate of metastatic progression under non-stress and stress conditions. (d) Ex vivo quantification of distant metastasis in target organs that were harvested from mice 28 days after tumor cell injection (n = 5). (e) Representative images of metastasis in lung and lymph node ex vivo at 28 days post tumor cell inoculation. Scale: lymph node: min. 5 × 10⁶; max. 1 × 10⁸ and lung: min. 2 × 10⁵; max. 8 × 10⁶ photons/sec. Data represent mean ± standard error. *p < 0.05 and ***p < 0.001 by (c) one-way ANOVA or two-way ANOVA (a) with or (d) without repeated measures followed by post hoc Tukey’s planned comparison tests.

Fig. 5. Stress-enhanced metastasis is independent of primary tumor size

(a) Mass of tumors derived from control or β₂AR-deficient cells, from mice exposed to non-stress or stress conditions (n = 5). (b – d) Pearson correlation analysis of the relationship between the effect of stress on the trajectory of metastasis and the effect of stress on the trajectory of primary tumor growth in mice bearing (b) control (r² = 0.001938, p = 0.944), (c) shADRB2a (r² = 0.9224, p = 0.0094), and (d) shADRB2b (r² = 0.6995, p = 0.0775) tumors. #p < 0.05 for main stress effect by two-way ANOVA.