Metformin disrupts crosstalk between G protein-coupled receptor and insulin receptor signaling systems and inhibits pancreatic cancer growth

Abstract

Recently, we identified a novel crosstalk between insulin and G protein-coupled receptor (GPCR) signaling pathways in human pancreatic cancer cells. Insulin enhanced GPCR signaling through a rapamycin-sensitive mTOR-dependent pathway. Metformin, the most widely used drug in the treatment of type 2 diabetes, activates AMP kinase (AMPK), which negatively regulates mTOR. Here, we determined whether metformin disrupts the crosstalk between insulin receptor and GPCR signaling in pancreatic cancer cells. Treatment of human pancreatic cancer cells (PANC-1, MIAPaCa-2, and BxPC-3) with insulin (10 ng/mL) for 5 minutes markedly enhanced the increase in intracellular [Ca(2+)] induced by GPCR agonists (e.g., neurotensin, bradykinin, and angiotensin II). Metformin pretreatment completely abrogated insulin-induced potentiation of Ca(2+) signaling but did not interfere with the effect of GPCR agonists alone. Insulin also enhanced GPCR agonist-induced growth, measured by DNA synthesis, and the number of cells cultured in adherent or nonadherent conditions. Low doses of metformin (0.1-0.5 mmol/L) blocked the stimulation of DNA synthesis, and the anchorage-dependent and anchorage-independent growth induced by insulin and GPCR agonists. Treatment with metformin induced striking and sustained increase in the phosphorylation of AMPK at Thr(172) and a selective AMPK inhibitor (compound C, at 5 micromol/L) reversed the effects of metformin on [Ca(2+)](i) and DNA synthesis, indicating that metformin acts through AMPK activation. In view of these results, we tested whether metformin inhibits pancreatic cancer growth. Administration of metformin significantly decreased the growth of MIAPaCa-2 and PANC-1 cells xenografted on the flank of nude mice. These results raise the possibility that metformin could be a potential candidate in novel treatment strategies for human pancreatic cancer.

Figures

Figure 1. Metformin blocks enhanced Ca2+ mobilization induced by neurotensin, bradykinin or angiotensin in insulin-treated human pancreatic cancer cells

A, Effect of metformin on Ca2+ mobilization induced by neurotensin in PANC-1 cells. PANC-1 cells were treated with or without 10 ng/ml insulin for 5 min and then stimulated with 5 nM neurotensin (NT) (A, left). [Ca2+]i was monitored as described in “Materials and Methods”. Pretreatment with 5mM metformin for 1 h prevented insulin-induced potentiation of the increase in [Ca2+]i (A, middle upper) but did not interfere with neurotensin-induced [Ca2+]i (A, middle lower). Right columns, Peak, Maximal increment in [Ca2+]i in response to 5 nM neurotensin (NT) in cells pretreated without or with 5 mM metformin (Met) for 1h and then incubated without or with 10 ng/ml insulin for 5 min (Ins) before NT stimulation. Results shown are the mean ± SEM; n = 25 for each condition. B, Effect of metformin on [Ca2+]i induced by neurotensin in MIAPaCa-2 cells and by bradykinin and angiotensin II in BxPC-3 cells. Left: MIAPaCa-2 cells were incubated for 1 h in the absence or presence of 5 mM metformin (Met) and then treated without or with 10 ng/ml insulin for 5 min (Ins) prior to stimulation with 5 nM neurotensin (NT). Right: BxPC-3 cells were incubated for 1 h in the absence or presence of 5 mM metformin (Met) and then treated without or with 10 ng/ml insulin for 5 min (Ins) prior to stimulation with 5 nM bradykinin (BK) or 100nM angiotensin II (AT). The values obtained with MIAPaCa-2 and BxPC-3 are the mean ± SEM of at least four independent experiments each performed by quadruplicate (n = 16 for each condition).

Figure 2. Metformin inhibits DNA synthesis induced by GPCR agonists and insulin in PANC-1, MIAPaCa-2 and BxPC-3 cells

Panels A, C and D: Dose-response effect of metformin on DNA synthesis induced by neurotensin, insulin or both in PANC-1 (A), MIAPaCa-2 (C) or BxPC-3 cells cells (D). PANC-1 and MIAPaCa-2 cells were incubated without (open bars, control) or with 5nM neurotensin (NT, grey bars), 10 ng/ml insulin (striped bars) or the combination of NT and insulin (black bars) in the presence of increasing concentration of metformin (0.5-5mM) for 17 h prior to the addition of [3H]-thymidine for 6 h. BxPC-3 cells were treated identically, except that they were stimulated with bradykinin instead of neurotensin. Values (radioactivity incorporated into acid-insoluble pools) are the mean ± SEM obtained in three independent experiments. Panel B: Metformin prevents prolonged ERK signaling. PANC-1 cells were incubated without or with 5 nM neurotensin (NT), 10 ng/ml insulin or the combination of NT and insulin in the absence or presence of 5 mM metformin (Met), as indicated. Cell lysates were subjected to Western Blot analysis using antibodies that detect dually phosphorylated ERK1/2 (pERK) or total ERK.

Figure 3. Metformin inhibits anchorage-dependent and independent proliferation of PANC-1 cells in response to neurotensin, insulin and their combination

Single-cell suspension of PANC-1 cells was plated either on tissue culture dishes (A) or on [Poly-(HEMA)]-coated dishes (B). After 24 h, the cultures were shifted to DMEM containing 1 % serum without (C) or with 5 nM neurotensin (NT), 10 ng/ml insulin (Ins), or a combination of NT and Ins in the absence (open bars) or presence (closed bars) of 1 mM metformin. The cultures were incubated for 7 days as described in “Materials and Methods”. Cell counts (determined from 4 to 6 wells per condition) are presented as mean ± SEM. Similar results were obtained in two independent experiments.

Figure 4. Metformin disrupts crosstalk between insulin receptor and GPCR signaling systems through AMPK

A, Metformin activates AMPK in PANC-1 cells. Lysates of PANC-1 cells treated with 5mM metformin for various time periods (0-24 h) were analyzed for AMPK phosphorylation at Thr172 as indicated in “Materials and Methods”. Shown here is a representative autoluminogram. Similar results were obtained in three independent experiments. B, AMPK inhibitor reverses the inhibitory effect of metformin on Ca2+ mobilization induced by neurotensin in PANC-1 cells. PANC-1 cells were pretreated with or without the AMPK inhibitor compound C (5μM) for 1h. Then, 5mM metformin was added to the media for 1 h and subsequently 10 ng/ml insulin for 5 min. The cells then were stimulated with 5 nM neurotensin and [Ca2+]i was monitored as described in “Materials and Methods”. Right columns, Peak, Maximal increment in [Ca2+]i in response to neurotensin in cells incubated without or with 10 ng/ml insulin for 5 min before stimulation with neurotensin (NT). Cultures were pretreated without or with 5 μM Compound C (AMPK inhib) for 1h followed by 5 mM metformin for 1h prior to the experiment (Met), as indicated. The values shown are the mean ± SEM; n = 25 derived from 5 independent experiments. We verified that metformin inhibited the enhanced increase in [Ca2+]i induced by neurotensin in insulin-treated cells (checked bar). C, The selective AMPK inhibitor Compound C reverses the inhibitory effect of metformin on DNA synthesis induced by neurotensin and insulin in PANC-1 cells. Cultures of PANC-1 cells were treated with or without 5 mM metformin (Met) and some of the cultures exposed to metformin were also treated with 5μM compound C (AMPK inhib). DNA synthesis was assayed as described in “Materials and Methods”.

Figure 5. Metformin inhibits the growth of PANC-1 and MIAPaCa-2 tumor xenografts

Xenografts were generated by implantation of 2×106 cells of PANC-1 (A) or MIAPaCa-2 (B) cells into the right flanks of male nu/nu mice. When the tumors reached a mean diameter of 2 mm the animals were randomized into control and treated groups (10 mice per group). Metformin (250 mg/kg) was given once daily intraperitoneally for the duration of the experiment (squares). The 1st day of treatment was designated as day 0. Control animals received saline. Tumor volumes were measured every 4 days as described in “Materials and Methods”. On day 41, PANC-1 tumors were removed, measured and tumor volumes estimated as V = 0.52 (length x width x depth). The results are shown in the inset (mean ± SEM). On day 25, the MIAPaCa-2 tumors were removed, measured and tumor volumes were estimated as described for PANC-1. The results are shown in the inset (mean ± SEM). *p<0.05, **p<0.01 versus control (Student's t test).