Altered serotonin physiology in human breast cancers favors paradoxical growth and cell survival

Abstract

Introduction: The breast microenvironment can either retard or accelerate the events associated with progression of latent cancers. However, the actions of local physiological mediators in the context of breast cancers are poorly understood. Serotonin (5-HT) is a critical local regulator of epithelial homeostasis in the breast and other organs. Herein, we report complex alterations in the intrinsic mammary gland serotonin system of human breast cancers.

Methods: Serotonin biosynthetic capacity was analyzed in human breast tumor tissue microarrays using immunohistochemistry for tryptophan hydroxylase 1 (TPH1). Serotonin receptors (5-HT1-7) were analyzed in human breast tumors using the Oncomine database. Serotonin receptor expression, signal transduction, and 5-HT effects on breast cancer cell phenotype were compared in non-transformed and transformed human breast cells.

Results: In the context of the normal mammary gland, 5-HT acts as a physiological regulator of lactation and involution, in part by favoring growth arrest and cell death. This tightly regulated 5-HT system is subverted in multiple ways in human breast cancers. Specifically, TPH1 expression undergoes a non-linear change during progression, with increased expression during malignant progression. Correspondingly, the tightly regulated pattern of 5-HT receptors becomes dysregulated in human breast cancer cells, resulting in both ectopic expression of some isoforms and suppression of others. The receptor expression change is accompanied by altered downstream signaling of 5-HT receptors in human breast cancer cells, resulting in resistance to 5-HT-induced apoptosis, and stimulated proliferation.

Conclusions: Our data constitutes the first report of direct involvement of 5-HT in human breast cancer. Increased 5-HT biosynthetic capacity accompanied by multiple changes in 5-HT receptor expression and signaling favor malignant progression of human breast cancer cells (for example, stimulated proliferation, inappropriate cell survival). This occurs through uncoupling of serotonin from the homeostatic regulatory mechanisms of the normal mammary epithelium. The findings open a new avenue for identification of diagnostic and prognostic markers, and valuable new therapeutic targets for managing breast cancer.

Figure 1

TPH1 expression in non-transformed and breast cancer cells. (A) Bar graphs representing TPH1 mRNA levels in indicated cell lines as detected by semi-quantitative RT - PCR reactions and normalized to GAPDH. Shown below is a representative picture of resulting PCR reactions. (B) Bar graphs and electophoretic bands showing TPH1 protein levels in the extracts of indicated cells normalized to β-actin. Error bars in (A) and (B) represent ± SEM. *P < 0.05, **P < 0.01 and ***P < 0.001 (one-way ANOVA) in comparison to MCF10A cells.

Figure 2

Changes in TPH1 signal in human breast tumors. (A) Representative image of normal human breast tissue (BT) from the tissue microarray stained for TPH1 (brown) and counterstained with hematoxylin. A corresponding no primary antibody control section is depicted on the bottom left. Magnified image shows details of epithelial staining and occasional positive cells in stroma. (B-D) Results depicting quantification of TPH1 immunostaining from blindly scored tissue microarray sections. The section scores for TPH1 were sorted as per human cancer cell line core (B), estrogen receptor (ER) status (C) and progesterone receptor (PR) status (D). TPH1 staining was reduced in ER negative tumors, but unaffected by PR status. Error bars represent ± SEM. *P < 0.05 in comparison with normal BT for (A).

Figure 3

Nonlinear association between TPH1 and cancer progression. (A-B) Results depicting quantification of TPH1 immunostaining from blindly scored tissue microarray sections. Data were sorted based on the indicated criteria. Data were analyzed by one-way ANOVA with P-values as noted in comparison with NI for (A) and N0 for (B). Error bars represent ± SEM. (A) TPH1 scores in samples identified as non-invasive (NI); invasive, node-negative (IN-); invasive, node-positive (IN+); and invasive with distant metastases (IN-Mets). (B) Human breast cancer tissue sections scored for TPH1 were sorted as per different stages of nodal metastasis (N-stage: N0, no regional lymph node involvement; N1, Mets to movable ipsilateral nodes; N1a/b, Mets up to 20 mm to 4 or more nodes with extension beyond node capsule; N2-3, Mets to ipsilateral nodes that are fixed to one another or to other structures and to internal mammary lymph nodes). (C) Representative microarray sections stained for TPH1 from sorted groups as indicated below each image. Brown/red dye indicates TPH1 stain, blue is hematoxylin.

Figure 4

Difference in 5-HT receptor isoform gene expression between non-transformed (pHMECs and MCF10A) and breast cancer cells (MCF7, MDA-MB-231 and T47D). (A) Representative RT-PCR reactions (mRNA transcripts) for indicated 5-HT receptors, in cells as indicated at the top. Human brain (hypothalamus) is used as a reference sample for each receptor. For primer information please see Table S1 in Additional data file 1. (B) Densitometric quantification (semi-quantitative) of transcript levels relative to human brain reference samples (100%). Each column represents average with error bars representing ± SEM of at least three independent reactions.

Figure 5

5-HT influence on apoptosis and morphological transition in breast cells. (A) Staining for apoptosis marker cleaved caspase-3 (green), counterstained for nuclei (blue) for the indicated cells with or without 5-HT (7.5 × 10^-4 M) treatment for 72 h in serum free media. (B) Phase-contrast images of the indicated cells with or without 5-HT (7.5 × 10^-4 M) treatment for 72 h. Arrowheads point to appendages at intersections between adjacent cells.

Figure 6

Growth inhibition by 5-HT in non-transformed human mammary epithelial cells is via 5-HT₇ recepto, r and breast cancer cells are refractory to growth inhibition by 5-HT. (Ai-iv) Proliferation assay (MTT) for detecting growth of primary human mammary epithelial cells (pHMECs) (Ai) and MCF10A (Aii-iv) after a period of 36 h following indicated treatments. Graphs are plotted as percent change from control (Ctrl - untreated) group. Error bars represent ± SEM. ***P < 0.001 (one-way ANOVA). (Ai and Aii), depict indicated cells treated with 5-HT (7.5 × 10^-4 M) in presence or absence of broad-spectrum 5-HT receptor antagonist methysergide (MS) (60 μM). (Aiii) depicts MCF10A cells treated with 5-HT (7.5 × 10^-4 M) in presence or absence of specific 5-HT₇ receptor antagonist SB 269970 (SB) (80 μM). (Aiv) depicts MCF10A cells treated with 5-HT (7.5 × 10^-4 M) in presence or absence of specific inhibitors of signaling proteins p38 MAPK (SB203580 [p38I]) (20 μM) and PKA (H89 [PKAI]) (10 μM). (Bi-ii) depict a proliferation assay (MTT-viable count) for detecting cell growth after 36 h of treatment with indicated 5-HT concentrations. Graphs are plotted as percent change from respective controls (untreated group). Error bars represent ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 (Two-way ANOVA). (Bi) depicts comparison of pHMECs (grey bars) and T47D (black bars) cells response to indicated 5-HT concentrations. (Bii) depicts comparison of MCF10A (grey bars) and MDA-MB-231 (black bars) cells response to indicated 5-HT concentrations. (Biii) depicts measurements of intracellular cAMP accumulation in MCF10A (grey squares and line) and MDA-MB-231 (black triangle and line) cells in response to 5-HT stimulation as indicated. Error bars in Biii represent ± SEM. ** P < 0.01, *** P < 0.001 (Two way ANOVA).

Figure 7

Model of a working hypothesis for 5-HT influences on breast cancer progression. 5-HT is a tumor-suppressing signal in non-transformed breast cells and early stage breast cancers, so early tumor growth relies on reductions in TPH1 expression. During tumor progression, cells acquire genetic or epigenetic alterations in 5-HT signaling which then make them resistant to suppressive 5-HT actions and favor tumor-promoting actions (e.g., dynamic cell junctions and cell shedding), as well as acquisition of new receptors and functions (e.g., stimulated proliferation and epithelial-mesenchymal transition). The figure is inspired by the image in [69].