Buformin inhibits the stemness of erbB-2-overexpressing breast cancer cells and premalignant mammary tissues of MMTV-erbB-2 transgenic mice

Abstract

Background: Metformin, an FDA-approved drug for the treatment of Type II diabetes, has emerged as a promising anti-cancer agent. Other biguanide analogs, including buformin and phenformin, are suggested to have similar properties. Although buformin was shown to reduce mammary tumor burden in carcinogen models, the anti-cancer effects of buformin on different breast cancer subtypes and the underlying mechanisms remain unclear. Therefore, we aimed to investigate the effects of buformin on erbB-2-overexpressing breast cancer with in vitro and in vivo models.

Methods: MTT, cell cycle, clonogenic/CFC, ALDEFLUOR, tumorsphere, and Western blot analyses were used to determine the effects of buformin on cell growth, stem cell populations, stem cell-like properties, and signaling pathways in SKBR3 and BT474 erbB-2-overexpressing breast cancer cell lines. A syngeneic tumor cell transplantation model inoculating MMTV-erbB-2 mice with 78617 mouse mammary tumor cells was used to study the effects of buformin (1.2 g buformin/kg chow) on tumor growth in vivo. MMTV-erbB-2 mice were also fed buformin for 10 weeks, followed by analysis of premalignant mammary tissues for changes in morphological development, mammary epithelial cell (MEC) populations, and signaling pathways.

Results: Buformin significantly inhibited SKBR3 and BT474 cell growth, and in vivo activity was demonstrated by considerable growth inhibition of syngeneic tumors derived from MMTV-erbB-2 mice. In particular, buformin suppressed stem cell populations and self-renewal in vitro, which was associated with inhibited receptor tyrosine kinase (RTK) and mTOR signaling. Consistent with in vitro data, buformin suppressed mammary morphogenesis and reduced cell proliferation in MMTV-erbB-2 mice. Importantly, buformin decreased MEC populations enriched with mammary reconstitution units (MRUs) and tumor-initiating cells (TICs) from MMTV-erbB-2 mice, as supported by impaired clonogenic and mammosphere formation in primary MECs. We further demonstrated that buformin-mediated in vivo inhibition of MEC stemness is associated with suppressed activation of mTOR, RTK, ER, and β-catenin signaling pathways.

Conclusions: Overall, our results provide evidence for buformin as an effective anti-cancer drug that selectively targets TICs, and present a novel prevention and/or treatment strategy for patients who are genetically predisposed to erbB-2-overexpressing breast cancer.

Keywords: Breast cancer; Buformin; Cancer stem cells; ErbB-2; Mammary epithelial cells.

Fig. 1

Buformin induces anti-proliferative effects in erbB-2-overexpressing SKBR3 and BT474 cells. a An MTT assay was performed to compare cell viability in SKBR3 and BT474 cells treated with buformin (0, 1, 3, 10, 30, 100, 300 μM, 1, 3, or 10 mM) for 5 days. The average viable cell fraction for each sample (N = 8) is plotted ± standard error (S.E.). b SKBR3 and BT474 cells were treated with buformin (0, 0.5, 1, or 3 mM) for 48 h and then cell cycle progression was measured using FACS analysis (N = 4). The average percentage of cells in G2/M, S, and G0/G1 phases of the cell cycle are graphed for SKBR3 (left panel) and BT474 (right panel) cells. c SKBR3 and BT474 cells were treated with buformin (0, 0.2, or 1 mM) in triplicate for 14 days as part of the clonogenic assay. Then the colonies were fixed and stained with crystal violet. The average number of colonies that formed after 14 days is graphed. Representative images of crystal violet-stained colonies are shown in the far right panel. Values are graphed as the mean ± S.E. (**p < 0.01)

Fig. 2

Buformin diet inhibits syngeneic tumor growth in MMTV-erbB-2 mice. 8-week-old MMTV-erbB-2 mice were inoculated with 78617 cells and tumors began to appear by Day 6 after the initial inoculation. At Day 6, mice (N = 7) were administered a buformin diet for 12 days with tumor palpitation every 2 days. The tumor volumes during the monitoring from Day 6 – 18 (a) and final tumor weights (b) were recorded. Representative images of tumors from the control and buformin-fed mice are shown in c. Values are graphed as the mean ± S.E. (**p < 0.01)

Fig. 3

Buformin suppresses the stemness of breast cancer cells in vitro. a SKBR3 and BT474 cells were treated with buformin (0, 0.5, or 1 mM) for 48 h (N = 4). Then ALDH activity was measured using an ALDEFLUOR assay. The percentage of ALDH⁺ cells are presented in the graphs. b 78617 and BT474 cells were plated in ultra-low attachment plates and treated with buformin (0, 0.2, or 0.5 mM) for 7 days (N = 4). Primary tumorspheres were counted and imaged after 7 days. Then the cells were harvested and homogenized to form a single cell suspension that was replated in ultra-low attachment plates for another 7 days. Secondary tumorspheres were then counted and imaged. All values are graphed as the mean ± S.E. (*p < 0.05, **p < 0.01)

Fig. 4

Buformin regulates AMPK/mTOR and RTK signaling pathways in erbB-2-overexpressing breast cancer cells. SKBR3 and BT474 cells were treated with buformin (0, 0.1, 0.3, 1, or 3 mM) for 24 h. Expression and activation/phosphorylation of the indicated proteins relating to the AMPK/mTOR (a) and RTK (b) pathways were analyzed using Western blotting

Fig. 5

Buformin diet during the premalignant risk window hinders mammary morphogenesis and proliferation in MMTV-erbB-2 mice. Mice were fed control and buformin diets for 10 weeks (N = 6). At 18 weeks of age, the mice were sacrificed for whole mount preparation of mammary glands (a). Representative images of H&E stained (b) and Ki67 immunostained (c) mammary tissues from 18-week-old mice are shown. Brown staining in C indicates Ki67⁺ cells, from which the percentage of Ki67⁺ cells from each group is graphed in d. All values are depicted in the graph as the mean ± S.E. (**p < 0.01)

Fig. 6

Buformin targets mammary stem cell populations in premalignant tissues from MMTV-erbB-2 mice. Primary MECs were isolated from 18-week-old mice that were fed control or buformin diets for 10 weeks (N = 4). Cells were labeled with fluorescent antibodies for flow cytometry analysis of different MEC populations. a The CD24/CD49f-probed cells produced 3 distinct cell populations, including luminal cells (left panel), myoepithelial cells (middle panel), and MRUs (right panel) that were compared between the control-fed and buformin-fed mice. Representative images of CD24/CD49f flow cytometry plots are shown in b. All values are depicted in the appropriate graphs as the mean ± S.E. (*p < 0.05, **p < 0.01)

Fig. 7

Buformin targets cancer stem cell/tumor-initiating cell populations in premalignant tissues from MMTV-erbB-2 mice. As detailed above, isolated MECs were labeled with CD61/CD49f markers and analyzed with flow cytometry (N = 4). a The percentage of CD61^highCD49f^mid cell populations were compared between the control and buformin-fed mice with representative CD61/CD49f flow cytometry plots shown in b. All values are depicted in the graph as the mean ± S.E. (**p < 0.01)

Fig. 8

Buformin impairs the stemness of primary MECs from MMTV-erbB-2 mice. a Primary MECs were plated for a CFC assay (N = 3). After 7 days, the cells were fixed and stained as previously described. The number of colonies in the control and buformin diet samples was recorded. b Primary MECs were used for a primary mammosphere assay where the isolated cells were plated in ultra-low attachment plates for 7 days as described previously (N = 3). Then the primary spheres were counted. All values are depicted in the appropriate graphs as the mean ± S.E. (**p < 0.01)

Fig. 9

Buformin regulates signaling pathways that promote cell proliferation. Protein lysates were extracted from premalignant mammary tissues of 18-week-old MMTV-erbB-2 mice that were fed control or buformin diets for 10 weeks (N = 3). The expression and activation/phosphorylation of proteins associated with the mTOR (a-b), RTK (c), ER (d), and Wnt/β-catenin (e) pathways were examined using Western blot analysis (a, c-e) and IHC (b). Representative images of p-mTOR and mTOR immunostained mammary tissues from 18-week-old mice are shown in b