Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 440, 57-68

Metformin Suppresses Adipogenesis Through Both AMP-activated Protein Kinase (AMPK)-dependent and AMPK-independent Mechanisms

Affiliations

Metformin Suppresses Adipogenesis Through Both AMP-activated Protein Kinase (AMPK)-dependent and AMPK-independent Mechanisms

Suet Ching Chen et al. Mol Cell Endocrinol.

Abstract

People with Type 2 diabetes mellitus (T2DM) have reduced bone mineral density and an increased risk of fractures due to altered mesenchymal stem cell (MSC) differentiation in the bone marrow. This leads to a shift in the balance of differentiation away from bone formation (osteogenesis) in favour of fat cell development (adipogenesis). The commonly used anti-diabetic drug, metformin, activates the osteogenic transcription factor Runt-related transcription factor 2 (Runx2), which may suppress adipogenesis, leading to improved bone health. Here we investigate the involvement of the metabolic enzyme, AMP-activated protein kinase (AMPK), in these protective actions of metformin. The anti-adipogenic actions of metformin were observed in multipotent C3H10T1/2 MSCs, in which metformin exerted reciprocal control over the activities of Runx2 and the adipogenic transcription factor, PPARγ, leading to suppression of adipogenesis. These effects appeared to be independent of AMPK activation but rather through the suppression of the mTOR/p70S6K signalling pathway. Basal AMPK and mTOR/p70S6K activity did appear to be required for adipogenesis, as demonstrated by the use of the AMPK inhibitor, compound C. This observation was further supported by using AMPK knockout mouse embryo fibroblasts (MEFs) where adipogenesis, as assessed by reduced lipid accumulation and expression of the adipogeneic transcription factor, C/EBPβ, was found to display an absolute requirement for AMPK. Further activation of AMPK in wild type MEFS, with either metformin or the AMPK-specific activator, A769662, was also associated with suppression of adipogenesis. It appears, therefore, that basal AMPK activity is required for adipogenesis and that metformin can inhibit adipogenesis through AMPK-dependent or -independent mechanisms, depending on the cellular context.

Keywords: AMPK; Adipogenesis; Diabetes; Mesenchymal stem cells; Metformin; Osteogenesis.

Figures

Fig. 1
Fig. 1
Metformin and the AMPK-activator, A769662, suppress adipogenesis of CH3H10T1/2 MSCs. A) Confluent CH3H10T1/2 MSCs were induced to differentiate by addition 10% foetal calf serum (FCS) in the presence or absence of adipogenic IID medium (insulin, isobutylmethylxanthine and dexamethasone) and/or 10 μM pioglitazone (PIO), 500 μM metformin or 100 μM A769662. After 5 days cells were fixed with formalin and stained with Oil Red O to detect neutral lipid accumulation. Representative micrographs from an experiment carried out on three separate occasions with similar results are shown. B) Confluent CH3H10T1/2 MSCs were induced to differentiate by addition 10% FCS in the presence or absence of IID medium and/or 500 μM metformin. Cell extracts were then prepared after 5 days and immunoblotted with antibodies to perilipin, adiponectin and tubulin. Representative immunoblots from an experiment carried out on three separate occasions with similar results are shown. Densitometric analysis of three immunoblots are shown as means ± SEM in the lower panel. Significant increases relative to control are indicated, *p < 0.05 and ***, p < 0.001.
Fig. 2
Fig. 2
Effects of Metformin and the AMPK-activator, A769662, on early markers of differentiation in C3H10T1/2 cells. A) Confluent CH3H10T1/2 cells were stimulated for 5 days with 10 μM pioglitazone (PIO) or an adipogenic (IID) or osteogenic (AGD) cocktail, in the presence or absence of 500 μM metformin or 100 μM A769662. Cell extracts were then prepared and immunoblotted with antibodies to PPARγ, Runx2 and AMPK. The phosphorylation-dependent electrophoretic mobility shift of Runx2 induced by AGD treatment is indicated (pRunx2). Representative immunoblots from an experiment carried out on three separate occasions with similar results are shown. B) Densitometric analysis of PPARγ1 (upper panel) and PPARγ2 (lower panel) levels relative to Runx2 are shown as means ± SEM. Significant increases (*, p < 0.05) relative to control, and significant decreases relative to PIO-stimulated cells (#, p < 0.05), are indicated (n = 3). Non-significant changes are also indicated (ns). C) Confluent C3H10T1/2 cells were transfected with a PPARγ (upper panel) and Runx2 (lower panel) luciferase gene reporter constructs, together with control Renilla luciferase vector, and then stimulated for two days with 500 μM metformin or 100 μM A769662, in the presence or absence of 10 μM pioglitazone (PIO), adipogenic medium (IID) or osteogenic medium (AGD). Cell extracts were then prepared and luciferase activities were measured using a dual luciferase reporter assay. Luciferase activities from three separate experiments are shown as means ± SEM. Significant increases in luciferase activity are indicated; *, p < 0.05, **, p < 0.01 and ***, p < 0.001, as are significant decreases in activity, #, p < 0.05 (n = 3). Non significance is also indicated (ns).
Fig. 3
Fig. 3
The AMPK activator, A769662, but not metformin activates AMPK in C3H10T1/2 cells. Confluent C3H10T1/2 cells were stimulated for the indicated times with 100 μM A769662 or 500 μM metformin. Cell extracts were then prepared and immunoblotted with antibodies towards the phosphorylated form of the AMPK substrate, ACC, or total ACC, as indicated in the upper panel. Densitometric values were obtained from immunoblots from three separate experiments and are shown as means ± SEM in the line graph in the lower panel. Significant increases in pACC are indicated; **, p < 0.01 (n = 3). Non significance is also indicated (ns).
Fig. 4
Fig. 4
The mTOR inhibitor, rapamycin, suppresses adipogenesis of C3H10T1/2 cells. A) Confluent CH3H10T1/2 cells were induced to differentiate by addition of 10% foetal calf serum (FCS) supplemented with adipogenic cocktail (IID), in the presence or absence of the mTOR inhibitor, 10 μM rapamycin. After 5 days cells were fixed with formalin and stained with Oil Red O to detect neutral lipid accumulation. Representative micrographs from an experiment carried out on three separate occasions with similar results are shown. B) Confluent C3H10T1/2 cells were transfected with PPARγ luciferase gene reporter construct, together with control Renilla luciferase vector and then stimulated for two days with 500 μM metformin or 10 μM rapamycin, in the presence or absence of IID. Cell extracts were then prepared and luciferase activity was measured using a dual luciferase reporter assay. Luciferase activities from three separate experiments are shown as means ± SEM. Significant increases in PPARγ activity are indicated ***, p < 0.001, as are significant decreases in PPARγ activity, #, p < 0.05, relative to IID-stimulated cells (n = 3).
Fig. 5
Fig. 5
Effects of rapamycin on AMPK and p70S6Kactivities in C3H10T1/2 cells. A) Confluent CH3H10T1/2 cells were stimulated for 5 days with IID, in the presence or absence of 500 μM metformin, 100 μM A769662, 10 μM rapamycin or the AMPK inhibitor, 10 μM compound C. Cell extracts were then prepared and immunoblotted with antibodies to phosphorylated ACC (Ser 79). Representative immunoblots from an experiment carried out on three separate occasions with similar results are shown (upper panel). Densitometric values from 3 separate experiments are shown in the lower panel as means ± SEM. Significant increases relative to control are indicated, **, p < 0.01 and ***, p < 0.001 (n = 3). B) Confluent CH310T1/2 cells were stimulated for 5 days with IID, in the presence or absence of 500 μM metformin, 100 μM A769662, 10 μM compound C or 10 μM rapamycin. Cell extracts were then prepared and immunoblotted with antibodies to phosphorylated p70S6K and total p70S6K. Representative immunoblots from an experiment carried out on three separate occasions with similar results are shown in the upper panel. Densitometric analysis of mean ± SEM p70S6K phosphorylation from 3 separate experiments are shown in the lower panel. Significant increases (*, p < 0.05) relative to control, and significant decreases relative to IID-stimulated cells (##, p < 0.01 and ###, p < 0.001), are indicated (n = 3).
Fig. 6
Fig. 6
Compound C inhibits adipogenesis of C3H10T1/2 cells. A) Confluent CH3H10T1/2 cells were stimulated for 5 days with IID, in the presence or absence of the AMPK inhibitor, 10 μM compound C. Cell extracts were then prepared and immunoblotted with antibodies to phosphorylated ACC (Ser 79). Representative immunoblots from an experiment carried out on three separate occasions with similar results are shown (upper panel). Densitometric values from 3 separate experiments are shown in the lower panel as means ± SEM. Significant decreases relative to control are indicated, #, p < 0.05. B) Confluent CH3H10T1/2 cells were treated with 10% FCS supplemented with adipogenic cocktail (IID), in the presence or absence of the indicated concentrations of compound C. After 5 days cells were fixed with formalin and stained with Oil Red O to detect neutral lipid accumulation. Representative micrographs from an experiment carried out on three separate occasions with similar results are shown. C) Confluent CH3H10T1/2 MSCs were induced to differentiate by addition 10% FCS in the presence or absence of IID medium and/or 10 μM Compound C or 10 μM rapamycin. Cell extracts were then prepared after 5 days and immunoblotted with antibodies to perilipin and tubulin. Representative immunoblots from an experiment carried out on three separate occasions with similar results are shown. Densitometric analysis of three immunoblots are shown as means ± SEM in the lower panel. Significant increases relative to control are indicated, ***p < 0.001 and significant decreases with respect to IID-treated cells are indicated, ###, p < 0.001.
Fig. 7
Fig. 7
Metformin suppresses adipogenesis in MEFs. A) Wild type and AMPK knockout (−/−) mouse embryonic fibroblasts (MEFs) were treated with IID plus 10 μM pioglitazone (PIO) for 7 or 9 days as indicated. Wild type MEFs were also stimulated in the presence or absence of 1 mM metformin or 100 μM A769662. Cells were then stained with Oil Red O. Representative micrographs from an experiment carried out on three separate occasions with similar results are shown. B) Confluent wild-type (upper panel) and AMPK−/− (lower panel) MEFs were stimulated for the indicated times with 10 μM pioglitazone (PIO) plus (IID) and, for wild-type MEFs, in the presence or absence of 1 mM metformin or 100 μM A769662. Cell extracts were then prepared and immunoblotted with antibodies to C/EBPβ and/or C/EBPδ, as indicated. Representative immunoblots from experiments carried out on three separate occasions with similar results are shown. C) Densitometric values taken at day 5 from the experiment carried out in Fig. 7B are shown in the lower panel as means ± SEM. Significant increases in expression are indicated, **, p < 0.01 and ***, p < 0.001. Significant decreases relative to control are also indicated, #, p < 0.05. Non-significance is also indicated (ns). D) Wild type MEFs were incubated with 1 mM metformin for the times indicated or with 100 μM A769662 for 30 min. Cells were then lysed and the lysates were immunoblotted with antibodies specific to the indicated proteins (upper panel). Results from densitometric analysis of three separate immunoblots are shown in the lower panel as means ± SEM. Significant increases relative to t = 0 are indicated, *, p < 0.05.

Similar articles

See all similar articles

Cited by 20 PubMed Central articles

See all "Cited by" articles

References

    1. Banerjee C., McCabe L.R., Choi J.Y., Hiebert S.W., Stein J.L., Stein G.S., Lian J.B. Runt homology domain proteins in osteoblast differentiation: AML3/CBFA1 is a major component of a bone-specific complex. J. Cell Biochem. 1997;66:1–8. - PubMed
    1. Bang S., Chen Y., Ahima R.S., Kim S.F. Convergence of IPMK and LKB1-AMPK signaling pathways on metformin action. Mol. Endocrinol. 2014;28:1186–1193. - PMC - PubMed
    1. Bergeron R., Ren J.M., Cadman K.S., Moore I.K., Perret P., Pypaert M., Young L.H., Semenkovich C.F., Shulman G.I. Chronic activation of AMP kinase results in NRF-1 activation and mitochondrial biogenesis. Am. J. Physiol. Endocrinol. Metab. 2001;281:E1340–E1346. - PubMed
    1. Cao Z., Umek R.M., McKnight S.L. Regulated expression of three C/EBP isoforms during adipose conversion of 3T3-L1 cells. Genes Dev. 1991;5:1538–1552. - PubMed
    1. Chen Q., Shou P., Zheng C., Jiang M., Cao G., Yang Q., Cao J., Xie N., Velletri T., Zhang X., Xu C., Zhang L., Yang H., Hou J., Wang Y., Shi Y. Fate decision of mesenchymal stem cells: adipocytes or osteoblasts? Cell Death Differ. 2016;0000023:1128–1139. - PMC - PubMed

MeSH terms

Feedback