Darko Stevanovic

Dual antiglioma action of metformin: cell cycle arrest and mitochondria-dependent apoptosis

Abstract.The present study reports for the first time a dual antiglioma effect of the well-known antidiabetic drug metformin. In low-density cultures of the C6 rat glioma cell line, metformin blocked the cell cycle progression in G0/G1 phase without inducing significant cell death. In confluent C6 cultures, on the other hand, metformin caused massive induction of caspase-dependent apoptosis associated with c-Jun N-terminal kinase (JNK) activation, mitochondrial depolarization and oxidative stress. Metformin-triggered apoptosis was completely prevented by agents that block mitochondrial permeability transition (cyclosporin A) and oxygen radical production (N-acetylcisteine), while the inhibitors of JNK activation (SP600125) or glycolysis (sodium fluoride, iodoacetate) provided partial protection. The antiglioma effect of metformin was reduced by compound C, an inhibitor of AMP-activated protein kinase (AMPK), and was mimicked by the AMPK agonist AICAR. Similar effects were observed in the human glioma cell line U251, while rat primary astrocytes were completely resistant to the antiproliferative and proapoptotic action of metformin.

In vitro and in vivo anti-melanoma action of metformin

The in vitro and in vivo anti-melanoma effect of antidiabetic drug metformin was investigated using B16 mouse melanoma cell line. Metformin caused a G(2)/M cell cycle arrest associated with apoptotic death of melanoma cells, as confirmed by the flow cytometric analysis of cell cycle/DNA fragmentation, phosphatidylserine exposure and caspase activation. Metformin-mediated apoptosis of melanoma cells was preceded by induction of oxidative stress and mitochondrial membrane depolarization, measured by flow cytometry in cells stained with appropriate fluorescent reporter dyes. The expression of tumor suppressor protein p53 was increased, while the mRNA levels of anti-apoptotic Bcl-2 were reduced by metformin, as revealed by cell-based ELISA and real-time RT-PCR, respectively. Treatment with metformin did not stimulate expression of the cycle blocker p21, indicating that p21 was dispensable for the observed cell cycle arrest. The activation of AMP-activated protein kinase (AMPK) was not required for the anti-melanoma action of metformin, as AMPK inhibitor compound C completely failed to restore viability of metformin-treated B16 cells. Metformin induced autophagy in B16 cells, as demonstrated by flow cytometry-detected increase in intracellular acidification and immunoblot-confirmed upregulation of autophagosome-associated LC3-II. Autophagy inhibitors ammonium chloride and wortmannin partly restored the viability of metformin-treated melanoma cells. Finally, oral administration of metformin led to a significant reduction in tumor size in a B16 mouse melanoma model. These data suggest that anti-melanoma effects of metformin are mediated through p21- and AMPK-independent cell cycle arrest, apoptosis and autophagy associated with p53/Bcl-2 modulation, mitochondrial damage and oxidative stress.

AMP-activated protein kinase-dependent and -independent mechanisms underlying in vitro antiglioma action of compound C.

We investigated the effect of compound C, a well-known inhibitor of the intracellular energy sensor AMP-activated protein kinase (AMPK), on proliferation and viability of human U251 and rat C6 glioma cell lines. Compound C caused G(2)/M cell cycle block, accompanied by apoptotic glioma cell death characterized by caspase activation, phosphatidylserine exposure and DNA fragmentation. The mechanisms underlying the pro-apoptotic action of compound C involved induction of oxidative stress and downregulation of antiapoptotic molecule Bcl-2, while no alteration of pro-apoptotic Bax was observed. Compound C diminished AMPK phosphorylation and enzymatic activity, resulting in reduced phosphorylation of its target acetyl CoA carboxylase. AMPK activators metformin and AICAR partly prevented the cell cycle block, oxidative stress and apoptosis induced by compound C. The small interfering RNA (siRNA) targeting of human AMPK mimicked compound C-induced G(2)/M cell cycle arrest, but failed to induce oxidative stress and apoptosis in U251 glioma cells. In conclusion, our data indicate that AMPK inhibition is required, but not sufficient for compound C-mediated apoptotic death of glioma cells.

Ghrelin-induced food intake and adiposity depend on central mTORC1/S6K1 signaling.

Signaling through the mammalian target of rapamycin complex 1 (mTORC1) and its effectors the S6-kinases (S6K) in the hypothalamus is thought to be involved in nutrient sensing and control of food intake. Given the anatomical proximity of this pathway to circuits for the hormone ghrelin, we investigated the potential role of the mTORC1/S6K pathway in mediating the metabolic effects of ghrelin. We found that ghrelin promoted phosphorylation of S6K1 in the mouse hypothalamic cell line N-41 and in the rat hypothalamus after intracerebroventricular administration. Rapamycin, an inhibitor of mTORC1, suppressed ghrelin-induced phosphorylation of hypothalamic S6K1 and increased food intake and insulin in rats. Chronic peripheral administration of ghrelin induced a significant increase in body weight, fat mass and food efficiency in wild-type and S6K2-knockout but not in S6K1-knockout mice. We therefore propose that ghrelin-induced hyperphagia, adiposity and insulin secretion are controlled by a central nervous system involving the mTORC1/S6K1 pathway.

Therapeutic improvement of glucoregulation in newly diagnosed type 2 diabetes patients is associated with a reduction of IL-17 levels.

We explored the effect of therapeutic glucoregulation on the blood levels of proinflammatory T helper (Th)17 cytokines interleukin (IL)-17 and IL-23, and Th1 cytokines interferon (IFN)-γ and IL-12 in newly diagnosed type 2 diabetes patients. The investigated group consisted of 23 subjects (17 men and 6 women, age 26-64). The cytokine serum levels, glycated hemoglobin (HbA1c) as a marker of glucoregulation, homeostasis model assessment index as a measure of insulin resistance (HOMA-IR), and body mass index (BMI) were determined before and after 12 weeks of therapy consisting of standard lifestyle modification and metformin (1000 mg b.i.d.). The levels of Th17 and Th1 cytokines before treatment did not correlate with age, BMI or HOMA-IR. The patients with poor glucoregulation (HbA1c>7%, n=12), compared to those with good glucoregulation (HbA1c≤7%, n=11), had higher serum levels of Th17 and Th1 cytokines, but only the differences in IL-17 (median 21.2 pg/ml vs. 4.8 pg/ml) and IFN-γ 5 (0.6 pg/ml vs. 27.7 pg/ml) reached statistical significance (p=0.003 and p=0.012, respectively). The reduction of HbA1c values (from 8.6 to 5.9%, p=0.000) observed upon treatment in patients with poor glucoregulation was associated with a significant decrease in the concentration of IL-17 (from 21.2 to 12.9 pg/ml, p=0.020), but not IFN-γ (50.6 vs. 52.3, p=0.349). These data indicate that therapeutic improvement of glucoregulation might contribute to a reduction of IL-17 levels in newly diagnosed type 2 diabetes patients.

Immunomodulatory actions of central ghrelin in diet-induced energy imbalance

We investigated the effects of centrally administered orexigenic hormone ghrelin on energy imbalance-induced inflammation. Rats were subjected for four weeks to three different dietary regimes: normal (standard food), high-fat (standard food with 30% lard) or food-restricted (70%, 50%, 40% and 40% of the expected food intake in 1st, 2nd, 3rd and 4th week, respectively). Compared to normal-weight controls, starved, but not obese rats had significantly higher levels of proinflammatory cytokines (TNF, IL-1β, IFN-γ) in the blood. When compared to normally fed animals, the hearts of starved and obese animals expressed higher levels of mRNAs encoding proinflammatory mediators (TNF, IL-1β, IL-6, IFN-γ, IL-17, IL-12, iNOS), while mRNA levels of the anti-inflammatory TGF-β remained unchanged. Intracerebroventricular (ICV) injection of ghrelin (1 μg/day) for five consecutive days significantly reduced TNF, IL-1β and IFN-γ levels in the blood of starved rats, as well as TNF, IL-17 and IL-12p40 mRNA expression in the hearts of obese rats. Conversely, ICV ghrelin increased the levels of IFN-γ, IL-17, IL-12p35 and IL-12p40 mRNA in the heart tissue of food-restricted animals. This was associated with an increase of immunosuppressive ACTH/corticosterone production in starved animals and a decrease of the immunostimulatory adipokine leptin both in food-restricted and high-fat groups. Ghrelin activated the energy sensor AMP-activated protein kinase (AMPK) in the hypothalamus and inhibited extracellular signal-regulated kinase (ERK) in the hearts of obese, but not starved rats. Therefore, central ghrelin may play a complex role in energy imbalance-induced inflammation by modulating HPA axis, leptin and AMPK/ERK signaling pathways.

Intracerebroventricular administration of metformin inhibits ghrelin-induced Hypothalamic AMP-kinase signalling and food intake.

Background/Aims: The antihyperglycaemic drug metformin reduces food consumption through mechanisms that are not fully elucidated. The present study investigated the effects of intracerebroventricular administration of metformin on food intake and hypothalamic appetite-regulating signalling pathways induced by the orexigenic peptide ghrelin. Methods: Rats were injected intracerebroventricularly with ghrelin (5 µg), metformin (50, 100 or 200 µg), 5-amino-imidazole-4-carboxamide 1-β-D-ribofuranoside (AICAR, 25 µg) and L-leucine (1 µg) in different combinations. Food intake was monitored during the next 4 h. Hypothalamic activation of AMP-activated protein kinase (AMPK), acetyl-CoA carboxylase (ACC), regulatory-associated protein of mTOR (Raptor), mammalian target of rapamycin (mTOR) and p70 S6 kinase 1 (S6K) after 1 h of treatment was analysed by immunoblotting. Results: Metformin suppressed the increase in food consumption induced by intracerebroventricular ghrelin in a dose-dependent manner. Ghrelin increased phosphorylation of hypothalamic AMPK and its targets ACC and Raptor, which was associated with the reduced phosphorylation of mTOR. The mTOR substrate, S6K, was activated by intracerebroventricular ghrelin despite the inhibition of mTOR. Metformin treatment blocked ghrelin-induced activation of hypothalamic AMPK/ACC/Raptor and restored mTOR activity without affecting S6K phosphorylation. Metformin also reduced food consumption induced by the AMPK activator AICAR while the ghrelin-triggered food intake was inhibited by the mTOR activator L-leucine. Conclusion: Metformin could reduce food intake by preventing ghrelin-induced AMPK signalling and mTOR inhibition in the hypotalamus.

Unacylated Ghrelin Suppresses Ghrelin-Induced Neuronal Activity in the Hypothalamus and Brainstem of Male Rats

Ghrelin, the endogenous growth hormone secretagogue, has an important role in metabolic homeostasis. It exists in two major molecular forms: acylated (AG) and unacylated (UAG). Many studies suggest different roles for these two forms of ghrelin in energy balance regulation. In the present study, we compared the effects of acute intracerebroventricular administration of AG, UAG and their combination (AG+UAG) to young adult Wistar rats on food intake and central melanocortin system modulation. Although UAG did not affect food intake it significantly increased the number of c-Fos positive neurons in the arcuate (ARC), paraventricular (PVN) and solitary tract (NTS) nuclei. In contrast, UAG suppressed AG-induced neuronal activity in PVN and NTS. Central UAG also modulated hypothalamic expression of Mc4r and Bmp8b, which were increased and Mc3r, Pomc, Agrp and Ucp2, which were decreased. Finally, UAG, AG and combination treatments caused activation of c-Fos in POMC expressing neurons in the arcuate, substantiating a physiologic effect of these peptides on the central melanocortin system. Together, these results demonstrate that UAG can act directly to increase neuronal activity in the hypothalamus and is able to counteract AG-induced neuronal activity in the PVN and NTS. UAG also modulates expression of members of the melanocortin signaling system in the hypothalamus. In the absence of an effect on energy intake, these findings indicate that UAG could affect energy homeostasis by modulation of the central melanocortin system.

Central Ghrelin Affects Pituitary-Thyroid Axis: Histomorphological and Hormonal Study in Rats

Body weight depends on the balance between energy intake and consumption. An interaction between ghrelin and thyroid function has been reported only in pathophysiological states. We examined whether intracerebroventricular (ICV) administration of ghrelin affects the structure and function of the pituitary-thyroid axis in young adult male rats. Ghrelin (0.3 nmol/5 μl PBS) or an equal volume of PBS were injected every 24 h into the lateral cerebral ventricle for 5 days. Two hours after the last treatment the animals were killed, their pituitaries and thyroids excised and prepared for further histological, immunohistochemical and morphometric investigation. Serum TSH levels were measured by RIA, while the total T4 and T3 levels were examined by ECLIA. Ghrelin treatment increased pituitary weight (p < 0.05) when compared to the controls, with no effect on the thyroid weight. Smaller, degranulated TSH-immunopositive cells were noticed within the pituitaries of ghrelin-treated animals; their cellular and nuclear volume as well as the relative volume density of thyrotrophs decreased (p < 0.05) in comparison to the control values. The level of serum TSH was reduced (p < 0.05). In the thyroid parenchyma of ghrelin-treated rats, an increased number of hypofunctioning follicles was noticed, characterized by flattened, weakly Tg-immunoreactive epithelium and colloid distension. The relative volume densities of the follicles and colloid increased (p < 0.05), while the thyroid index of activation rate and the serum level of total T4 decreased (p < 0.05). In conclusion, centrally applied ghrelin modulated the immunohistomorphometric features of pituitary TSH cells and decreased the level of serum TSH, consequently changing thyroid morphology and function, by reducing the T4 hormone level in the serum.

Consummatory behavior and metabolic indicators after central ghrelin injections in rats

Ghrelin, an endogenous ligand for the growth-hormone-secretagogue receptor, is a 28-amino acid peptide with a post-translational acyl modification necessary for its activity. It has central nervous system actions that affect appetite, body mass and energy balance. An intracerebroventricular (ICV) injection protocol of sub-nanomolar doses of ghrelin, known to alter the morphology of ACTH and GH producing pituicytes and plasma levels of these hormones, was used to provide an overview of metabolic changes linked to energy metabolism. Variables measured were: food intake (FI), water intake (WI), fecal mass, urine volume, body weight (BW), retroperitoneal (RP) and epididymal (EPI) white adipose tissue (WAT), and changes in serum leptin, insulin, triglycerides, cholesterol, and glucose. Five injections of rat ghrelin or PBS (n=8 per group) were given ICV every 24 h (1 microg/5 muL PBS) to adult male rats. Ghrelin had a positive and cumulative effect on FI, WI and BW (p<0.05), but not feces mass or urine volume (p>0.05). Centrally applied ghrelin clearly increased RP WAT (by 235%, p<0.001), EPI WAT (by 85%, p<0.05) and serum insulin levels (by 43%, p<0.05), and decreased serum leptin levels (by 77%, p<0.05) without (p>0.05) evoking changes in blood triglyceride cholesterol, or glucose levels. These data and the available literature clearly document that exposure of the brain of normal rats, over time, to sub-nanomolar doses of ghrelin results in metabolic dysregulation culminating in increased body mass, consummatory behavior, and lipid stores as well as changes in blood leptin/insulin levels. Thus, modulation of central ghrelin receptors may represent a pharmacological approach for controlling multiple factors involved in energy balance and obesity.