Metformin targets the metabolic achilles heel of human pancreatic cancer stem cells

Abstract

Pancreatic ductal adenocarcinomas contain a subset of exclusively tumorigenic cancer stem cells (CSCs), which are capable of repopulating the entire heterogeneous cancer cell populations and are highly resistant to standard chemotherapy. Here we demonstrate that metformin selectively ablated pancreatic CSCs as evidenced by diminished expression of pluripotency-associated genes and CSC-associated surface markers. Subsequently, the ability of metformin-treated CSCs to clonally expand in vitro was irreversibly abrogated by inducing apoptosis. In contrast, non-CSCs preferentially responded by cell cycle arrest, but were not eliminated by metformin treatment. Mechanistically, metformin increased reactive oxygen species production in CSC and reduced their mitochondrial transmembrane potential. The subsequent induction of lethal energy crisis in CSCs was independent of AMPK/mTOR. Finally, in primary cancer tissue xenograft models metformin effectively reduced tumor burden and prevented disease progression; if combined with a stroma-targeting smoothened inhibitor for enhanced tissue penetration, while gemcitabine actually appeared dispensable.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Metformin targets pancreatic cancer stem cells.

(A) Primary PDAC cells, but not normal pancreas cells express organic cation transporter 1, 2, and 3 (n = 3). (B) Definition of the therapeutic range for metformin in primary PDAC cells. Number of cells grown in the presence of the indicated concentrations of metformin for 24 h (n = 6). (C) qPCR analysis of CSCs-associated genes in spheres treated with 3 mM of metformin for 7 days. Data are normalized to the housekeeping gene and are presented as fold change in gene expression relative to control cells (n = 6). (D) Representative Western blot illustrating reduced Oct4 protein expression in response to metformin treatment (n = 3). (E) Representative flow cytometry analysis for CSCs markers in spheres treated for 7 days with 3 mM of metformin as compared to untreated spheres (upper panel). Summary of data for PDAC-185, A6L, 215, 253, and 354 is shown (lower panel; n = 6).

Figure 2. Metformin diminishes in vitro and in vivo tumorigenicity.

(A) Metformin decreases the size of spheres. Representative images of spheres obtained after treatment with metformin for 7 days. Quantification of sphere size (n≥6). (B) Sphere formation capacity in the presence or absence of metformin for 7 days (n≥6). (C) Self-renewal capacity of cancer stem cells isolated from tumors responding poorly in terms of first passage sphere forming capacity. Cells were continuously passage as secondary and tertiary spheres treated with metformin or vehicle only during first generation sphere formation (n = 6). (D) Colony formation for PDAC-185, A6L, 215, 253, and 354 evidenced by 0.05% crystal violet after 21 days (n = 3). (E) Rarefication of in vivo tumorigenic cancer stem cells in spheres treated with metformin as compared to vehicle. (F) Invasion of sphere-derived cells after 24 h of treatment with metformin or control (n = 3).

Figure 3. Metformin specifically eliminates cancer stem cells.

(A) Number of cells grown in the presence of the indicated concentrations of metformin for 24 h. (n = 3). (B) Quantification for Ki67 and DAPI in adherent cells after 7 d of treatment with metformin or control (n = 3). (C) qPCR analysis for CyclinD1 in adherent and sphere-derived cells after 7 d of treatment with metformin or control. Data are normalized to the housekeeping gene and are presented as fold change in gene expression relative to untreated cells (n = 6). (D) Cell cycle analysis determined by Propidium Iodide staining in adherent cells and spheres after 7 d of treatment with metformin or control (n = 3). (E) Cytometry analysis of apoptotic cells by double staining for Annexin V/DAPI after treatment with metformin or control for adherent versus sphere-derived cells (n = 3).

Figure 4. Mechanism of action.

(A) Cellular ATP levels in adherent cells and spheres after 7 d of treatment with metformin or control (n = 3). (B) Upper panel: Western blot analysis of pAMPK, AMPK, and GAPDH in spheres treated with metformin or control. Lower panel: Western blot analysis of pAMPK, pp70S6K, and GAPDH in adherent cells and spheres treated with metformin or control for 7 days (n = 3). (C) Overall effect of metformin (3 mM), inhibition of mTOR (rapamycin; 100 ng/ml), and direction activation of AMPK (A769662; 10 µM) on sphere formation and (n = 6). (D) colony formation for PDAC-215, 253, and 354 cells (n = 3). (E) Total and mitochondrial (Mito) ROS production after 8 hours of control or metformin treatment. (F) Mitochondrial transmembrane potential after 8 hours of control or metformin treatment.

Figure 5. Metformin stalls PDAC progression in vivo.

(A) Left panel: PDAC-215 tissue was implanted and treatment was allocated after initial tumor take was verified. Mice were treated with either gemcitabine (Gem) or metformin (Met). Metformin was discontinued on d100 to test the potential of the remaining lesions to initiate disease relapse. After documented disease relapse, metformin treatment was re-administered. Right panel: Body weight during treatment. (B) Left panel: Histological analysis: Hematoxylin & Eosin (H&E), CyclinD1, and Caspase3. Right panel: Content for CD44+ cells. (C) Effects of in vitro treatment as indicated on sphere formation capacity. (D) PDAC-A6L tissue implanted in mice and treated as indicated. (E) Content for CD44+ cells (upper panel) and sphere formation capacity (lower panel). (F) Histological analysis: H&E and cytokeratin19 (CK19; n = 6).

Figure 6. Additional targeting of the stroma prevents tumor relapse.

(A) PDAC-185 tissue implanted in mice and treated as indicated including the smoothened inhibitor SIBI-C1. (B) Content for EpCAM+ and CD133+ cells, respectively, (upper panel) and sphere formation capacity (lower panel). (C) Histological analysis: Hematoxylin&Eosin (H&E) and cytokeratin19 (CK19). (D) mTOR inhibitor resistant PDAC-253 tissue implanted in mice and treated as indicated (n = 6).