Insulin activity plays an important role in glucose and lipid metabolism, and insulin resistance in muscle, liver, and adipose tissue is a significant threat to public health. This condition is thought to result from several physiological mechanisms, including but not limited to inflammation, ectopic lipid accumulation in insulin-responsive tissues, and the unfolded protein response (1). PPARγ and mTOR are 2 major regulators of insulin sensitivity, and gaining greater insight into their biology in relation to insulin resistance could yield better therapies for a number of disorders, including Type II diabetes.
The nuclear hormone receptor PPARγ, a master regulator of adipogenesis which is also involved in inflammation and cancer, is a major therapeutic target in insulin resistance. Stimulation of PPARγ with TZDs, PPARγ agonists, is effective for restoring insulin sensitivity in adipose tissue (2), but TZDs can also lead to several negative side effects, including weight gain, edema, and congestive heart failure (3). The Ser/Thr protein kinase mechanistic target of rapamycin (mTOR) also plays a significant role in regulating insulin signaling. In combination with several other molecules, mTOR can form 2 complexes, mTORC1 and mTORC2. mTORC1 contains raptor (regulatory associated protein of Tor), whereas mTORC2 contains rictor (rapamycin-insensitive companion of Tor). Each complex plays a distinct role in regulating insulin sensitivity: mTORC1 inhibits insulin signaling via its substrate S6K1, whereas mTORC2 has been shown to have a positive effect on glucose uptake and tolerance (4).
Recently, CDK5-mediated phosphorylation of PPARγ ser-273 has become a topic of therapeutic interest, and a recent study by Li et al reveals the role of NCoR in inhibiting adipocyte PPARγ activity by promoting CDK5-PPARγ binding (5). In addition to clarifying an aspect of PPARγ biology, the authors suggest that these findings may point to a strategy for avoiding some of the negative side effects of TZD treatment (see figure Insulin resistance
The authors made an adipocyte-specific NCoR knockout mouse (AKO) and showed that on a high-fat diet (HFD), AKO mice increased both food intake and obesity. However, insulin sensitivity improved in AKO muscle, liver, and adipose tissue. Intriguingly, though adipocyte hypertrophy is usually the mechanism for obesity, AKO mice appeared to display less hypertrophy and more hyperplasia than wild-types, as well as higher PPARγ levels. Finally, inflammation was also diminished in the AKO mice, characterized by lower levels of infiltrating adipose tissue macrophages and proinflammatory cytokine gene expression. All of these data suggested that NCoR may play a role in repressing PPARγ and allowing insulin resistance and inflammation to develop under high-fat diet conditions and obesity.
To unravel the mechanism for the improved insulin sensitivity and lowered inflammation in AKO mice, the team examined adipose tissue and found that levels of PPARγ phosphorylated at serine 273 were lower in NCoR knockouts. CDK5 phosphorylates PPARγ ser-273, blocking PPARγ transcriptional activity. Therefore, they assessed the physical interactions of NCoR, CDK5, and PPARγ, and found that NCoR is responsible for promoting the direct physical interaction of CDK5 and PPARγ. Rosiglitazone ablates NCoR - PPARγ interaction, and therefore diminishes CDK5's PPARγ association, and this association is also diminished in AKO. HFD-fed AKO mice also showed greater gene expression in the PPARγ signaling pathway and insulin signaling compared to WT. In summary, NCoR blocks PPARγ activity by driving CDK5 phosphorylation of ser-273, promoting insulin resistance.
The authors note that hemodilution, a side effect of TZDs, is absent from the AKO mice, indicating that adipocyte-specific targeting of NCoR may be an improvement on traditional TZD treatment. Indeed, a recent paper by Choi et al reported the development of a compound that blocked CDK5-mediated PPARγ phosphorylation by non-agonistic binding and produced antidiabetic effects without some of the negative side effects of TZDs (6). These studies indicate that targeting of the CDK5 phosphorylation event, rather than direct PPARγ agonism, may be a promising route forward in treatment of insulin resistance.
PPARγ transcriptional activity can also be modulated by another post-translational modification, sumoylation at K107. Fibroblast growth factor 21 (FGF21) is a circulating factor produced by the liver in response to fasting, and improves insulin sensitivity when administered to insulin-resistant animals. FGF21 is upregulated in response to TZD treatment in adipocytes, where it plays roles in differentiation and glucose uptake. A new study reveals a mechanism for these observations, showing that FGF21 is involved in preventing PPARγ K107 sumoylation in adipocytes, enhancing PPARγ activity and adipogenesis (7). Accordingly, FGF21 appears to play a key role in the mechanism of insulin sensitization via TZD treatment.
In striking contrast to FGF21's kinetics in the liver, the team found that FGF21 was produced in response to feeding in white adipocyte tissue (WAT), but not in circulation. Interestingly, FGF21 KO mice had less total fat mass while retaining approximately the same number of adipocytes. In vitro experiments with knockout cells suggested that FGF21 is not only important for adipocyte differentiation, but also for normal lipid accumulation in adipocytes. The team noted that rosiglitazone-induced differentiation and lipid accumulation was also diminished in FGF21 knockouts, and was restored with FGF21 treatment. FGF21 knockouts also had higher levels of sumoylated PPARγ, and further study using mutations at PPARγ's sumoylation sites revealed that this modification took place at K107, which is known to block PPARγ activity.
The team next sought to define FGF21's role in the response to rosiglitazone treatment. They found that diet-induced obese FGF21 knockout mice did not experience many of the same benefits as DIO wild-type, such as improved glucose disposal in glucose and insulin tolerance tests or significant decreases in plasma glucose or insulin. However, some parameters did not change between WT and KO, suggesting that FGF21 is only necessary for some of rosiglitazone's functions. Interestingly, increased fat mass and fluid mass, common negative side effects of TZD treatment, were also not observed in the knockouts, indicating that FGF21 may play a role in these effects as well.
In addition to recent strides in understanding the mechanism of PPARγ activity, a new study from Zhang et al sheds light on how triacylglycerol (TAG) synthesis leads to insulin resistance in the liver by way of mTORC2 (8). mTORC2 is the complex of mTOR and rictor (rapamycin-insensitive companion of TOR), and in contrast to mTORC1, it is known to increase insulin signaling through phosphorylation of Akt.
In this study, the authors initially demonstrated that overexpression of a TAG synthesis enzyme, GPAT1, suppressed insulin signaling and insulin's ability to repress gluconeogenesis in hepatocytes. mTOR - rictor binding was also diminished, leading to less mTORC2 phosphorylation of Akt-S473 and NDRG1. Conversely, mTORC1 signaling seemed to be unaffected. The importance of GPAT1 in influencing mTORC2 formation and activity was confirmed using GPAT1 knockout hepatocytes, which showed greater phosphorylation of Akt following insulin stimulation, as well as increased association of the mTORC2 components mTOR and rictor.
Because GPAT1 initiates TAG synthesis, the team hypothesized that lipid intermediates were responsible for the inhibition of mTORC2 formation and activity. Indeed, both GPAT1 and AGPAT2 led to higher levels of phosphatidic acid (PA) and diacylglycerol (DAG) containing 16:0. Treatment with di-16:0-PA alone, but not other PA species or DAG, caused mTORC2 to dissociate. These findings suggest that mTOR is regulated by PA, leading to diminished mTORC2 and insulin signaling. Therefore, 16:0-containing PA may play an important role in generating insulin resistance in response to nutrient excess.
mTORC2 was also found to be crucial for glucose homeostasis and insulin sensitivity in the context of long-term rapamycin treatment (9). Rapamycin has been shown to improve lifespan in many organisms, similar to deletion of the mTORC1 substrate S6K1 (which is also known to inhibit insulin signaling, as mentioned previously) and calorie restriction. However, while these treatments improve insulin sensitivity, rapamycin treatment leads to insulin resistance.
Lamming and colleagues, working in mice and rats, observed insulin resistance in the liver in the context of chronic rapamycin treatment as well as higher expression levels of gluconeogenic genes, but normal glucose uptake in muscle and adipose tissue. While impairing mTORC1 signaling had no effect on this response, the team observed less activation of several mTORC2 substrates, and immunoprecipitation showed that both mTORC1 and mTORC2 were disrupted in liver, adipose tissue, and skeletal muscle. The team deleted Rictor to more specifically pinpoint the effects of losing mTORC2, and showed that deletion alone was enough to copy the glucose intolerance and upregulation of gluconeogenic genes observed with rapamycin; these effects were not amplified with rapamycin treatment in Rictor-deleted mice.
The authors finally used mice with impaired mTORC1 signaling to show that mTORC1 disruption led to longevity even when mTORC2 signaling and glucose tolerance were normal, demonstrating that the longevity benefits of rapamycin treatment are not inextricable from the undesirable metabolic side effects.
As knowledge continues to deepen regarding the mechanisms behind insulin resistance, the pool of potential therapeutic targets becomes more diverse. TZD drugs enhance insulin sensitivity through PPARγ agonism, but also trigger unwanted side effects such as weight gain and edema. The findings in NCoR and FGF21 knockout mice may help pinpoint which pathways are responsible for which TZD effects, and lend crucial insight into the mechanisms of insulin resistance and sensitization through PPARγ. The recent findings regarding mTORC2 disruption expand current knowledge about how TAG synthesis intermediates and rapamycin contribute to metabolic dysfunction, perhaps representing a step forward in discovering how to prevent development of insulin resistance and age-related diseases.
In order to continue dissecting all of the biological pieces that lead to insulin resistance, gaining a complete picture of gene expression under various conditions is a highly useful strategy. RT² Profiler PCR Arrays provide gene expression analysis for the 84 most relevant genes in your system, as well as custom arrays, giving a total picture of the changes in individual genes and their networks.
- Samuel, V.T., and Shulman, G.I. (2012) Mechanisms for insulin resistance: common threads and missing links. Cell 148, 852.
- Lehrke, M., and Lazar, M. (2005) The many faces of PPARγ. Cell 123, 993.
- Nesto, R.W. et al. (2004) Thiazolidinedione use, fluid retention, and congestive heart failure: a consensus statement from the American Heart Association and American Diabetes Association. Diabetes Care 27, 256.
- Laplante, M., Sabatini, D.M. (2012) mTOR signaling in growth control and disease. Cell 149, 274.
- Li, P. et al. (2011) Adipocyte NCoR knockout decreases PPARγ phosphorylation and enhances PPARγ activity and insulin sensitivity. Cell 147, 815.
- Choi, J.H. et al. (2011) Antidiabetic actions of a non-agonist PPARγ ligand blocking Cdk5-mediated phosphorylation. Nature 477, 477.
- Dutchak, P.A., Katafuchi, T., Bookout, A.L., Choi, J.H., Yu, R.T., Mangelsdorf, D.J., and Kliewer, S.A. (2012) Fibroblast growth factor-21 regulates PPARγ activity and the antidiabetic actions of thiazolidinediones. Cell 148, 556.
- Zhang, C., Wendel, A.A., Keogh, M.R., Harris, T.E., Chen, J., and Coleman, R.A. (2012) Glycerolipid signals alter mTOR complex 2 (mTORC2) to diminish insulin signaling. Proc. Natl. Acad. Sci. 109, 1667.
- Lamming, D.W. et al. (2012) Rapamycin-induced insulin resistance is mediated by mTORC2 loss and uncoupled from longevity. Science 335, 1638.