Salicylate suppresses the oncogenic hyaluronan network in metastatic breast cancer cells

Abstract

The oncogenic role of hyaluronan in several aspects of tumor biology has been well established. Recent studies by us and others suggest that inhibition of hyaluronan synthesis could represent an emerging therapeutic approach with significant clinical relevance in controlling different breast cancer subtypes, including triple-negative breast cancer. Epidemiological and preclinical studies have revealed the therapeutic potential of aspirin (acetyl salicylate), a classical anti-inflammatory drug, in patients with cancer. However, the underlying molecular mechanisms remain unknown. The present study demonstrates that salicylate, a break down product of aspirin in vivo, alters the organization of hyaluronan matrices by affecting the expression levels of hyaluronan synthesizing (HAS1, 2, 3) and degrading (HYAL-1, -2) enzymes, and that of hyaluronan receptor CD44. In particular, salicylate was found to potently activate AMPK, a kinase known to inhibit HAS2 activity, and caused a dose-dependent decrease of cell associated (intracellular and membrane-bound) as well as secreted hyaluronan, followed by the down-regulation of HAS2 and the induction of HYAL-2 and CD44 in metastatic breast cancer cells. These salicylate-mediated effects were associated with the redistribution of CD44 and actin cytoskeleton that resulted in a less motile cell phenotype. Interestingly, salicylate inhibited metastatic breast cancer cell proliferation and growth by inducing cell growth arrest without signs of apoptosis as evidenced by the substantial decrease of cyclin D1 protein and the absence of cleaved caspase-3, respectively. Collectively, our study offers a possible direction for the development of new matrix-based targeted treatments of metastatic breast cancer subtypes via inhibition of hyaluronan, a pro-angiogenic, pro-inflammatory and tumor promoting glycosaminoglycan.

Keywords: AMPK; Aspirin; Breast cancer; CD44; Hyaluronan; Hyaluronan synthase 2; Salicylate.

Fig. 1

Salicylate activates AMPK in metastatic breast cancer cells. Immunoblot analyses (n = 4) of phosphorylated AMPK at Thr¹⁷² in the α subunit and total AMPKα after (A) treatment with AICAR (1 mM), salicylate (10 mM) and 4-MU (1 mM) for 24 h, and (B) treatment with salicylate (10 mM) for various time periods (0 min, 5 min, 30 min, 3 h, 6 h, 12 h, 24 h), in the absence (0%) or presence (10%) of serum (FBS). Statistical differences with untreated cells are indicated with black asterisks (***p < 0.001).

Fig. 2

Salicylate inhibits hyaluronan biosynthesis and secretion in metastatic breast cancer cells. (A) Immunofluorescence analysis of intracellular and membrane-bound hyaluronan was performed with biotin-HABP (green) in MDA-MB-231 cells treated for 24 h with PBS (0 mM, control) or salicylate (10 mM) in the absence (0%) or presence (10%) of serum (FBS). Nuclei are shown in blue (DAPI). Scale bars ~40 μm. (B) Quantification of secreted hyaluronan amounts by a microtiter-based assay in conditioned media of MDA-MB-231 breast cancer cells treated for 6, 12 and 24 h with salicylate (5, 10 and 20 mM) in the absence (0%) or presence (10%) of serum (FBS). The values represent the mean ± SD of 3 independent experiments run in triplicate. Statistical differences (*p < 0.05, **p < 0.01, ***p < 0.001) between salicylate-treated and control (0 mM) cells, and between different treatments are indicated with black and red asterisks, respectively. Statistical differences between serum-starved cells (0% FBS) and cells cultured in the presence of serum (10% FBS) are indicated with hashtag (^#p < 0.001).

Fig. 3

Salicylate affects the expression of hyaluronan metabolizing enzymes (HASs, HYALs) and CD44 receptors in metastatic breast cancer cells. (A) Quantitative qPCR analyses of HAS1, HAS2, HAS3, HYAL-1, HYAL-2, CD44s, CD44v3, CD44v6 and CD44v9 after treatment with 10 mM salicylate for 24 h in the absence (0%) of serum (FBS). (B) Quantitative qPCR analysis of HAS2 expression after treatment with 10 mM salicylate for 6, 12 and 24 h in the absence (0%) or presence (10%) of serum (FBS). The values represent the mean ± SD of 3 independent experiments run in triplicate. Statistical differences (*p < 0.05, **p < 0.01, ***p < 0.001) between salicylate-treated and control (0 mM) cells, and between different treatments are indicated with black and red asterisks, respectively.

Fig. 4

Salicylate suppresses metastatic breast cancer cell viability and growth. Cell viability after treatment with PBS (0 mM, control) or increasing concentrations of salicylate (5, 10 and 20 mM) in the absence (0%) or presence (10%) of serum (FBS) for (A) 48 h (short exposure) and (B) 2, 4, 6, 8 and 10 days (long exposure). The values represent the mean ± SD of 3 independent experiments run in triplicate. (C) Immunoblot analyses (n = 3) of cyclin D1, total and cleaved caspase-3, and GAPDH in control or salicylate-treated MDA-MB-231 cells in the absence (0%) or presence (10%) of serum (24 h post-treatment). Statistical differences (**p < 0.01, ***p < 0.001) between salicylate-treated and control (0 mM) cells, and between different treatments are indicated with black and red asterisks, respectively. Statistical differences between serum-starved cells (0% FBS) and cells cultured in the presence of serum (10% FBS) are indicated with hashtag (^#p < 0.001).

Fig. 5

Salicylate induces changes in actin cytoskeleton and CD44 distribution, and inhibits the migration of metastatic breast cancer cells. (A) Migration of MDA-MB-231 cells treated with PBS (0 mM, control) or increasing concentrations of salicylate (5, 10 and 20 mM) in the absence (0%) or presence (10%) of serum (FBS). The values represent the mean ± SD of 4 independent experiments run in triplicate. Statistical differences between salicylate-treated and control cells are indicated with black asterisks (*p < 0.05, **p < 0.01, ***p < 0.001). (B) Immunofluorescence analysis for filamentous actin (F-actin, green) and CD44 (red) in MDA-MB-231 cells exposed to PBS or 10 mM salicylate for 24 h in the absence (0%) or presence (10%) of serum (FBS). Nuclei are shown in blue (DAPI). Scale bars ~40 μm.

Fig. 6

Salicylate enhances the adhesiveness of metastatic breast cancer cells. (A) Adhesion of MDA-MB-231 cells on collagen type I matrix after treatment with PBS (0 mM, control) or increasing concentrations of salicylate (5, 10 and 20 mM) for 24 h in the absence (0%) or presence (10%) of serum. (B) Immunofluorescence analysis of cells treated for 24 h with PBS or 10 mM salicylate for filamentous actin (F-actin, green) and CD44 (red) in the absence (0%) or presence (10%) of serum. Nuclei are shown in blue (DAPI). Arrows point at newly formed cell-cell contacts. Scale bars ~40 μm. (C) Quantitative qPCR analysis of E-cadherin expression after salicylate (10 mM) treatment for 24 h in the absence (0%) or presence (10%) of serum. The values represent the mean ± SD of 3 independent experiments run in triplicate. Statistical differences between salicylate-treated and control (0 mM) cells are indicated with black asterisks (*p ≤ 0.05, **p ≤ 0.01). Statistical differences between serum-starved cells (0% FBS) and cells cultured in the presence of serum (10% FBS) are indicated with hashtag (^#p < 0.001).

Fig. 7

Proposed model of salicylate-mediated onco-suppressive effects in metastatic breast cancer cells. Salicylate inhibits hyaluronan biosynthesis and accumulation through: (1) phosphorylation and activation of AMPK, (2) conjugation with precursor substrates (UDP-GlcUA), (3) down-regulation of HAS2 (in nutrient deprivation conditions) and up-regulation of HYAL-2. These changes are followed by the repression of cyclin D1 and the marked redistribution of actin filaments and CD44 resulting in inhibition of cell proliferation/growth and cell motility, respectively. For details, see text.