protein kinase (AMPK) and the histone/protein deacetylase SIRT1 are fuel-sensing molecules that have coexisted in cells throughout evolution. When a cell's energy state is diminished, AMPK activation restores energy balance by stimulating catabolic processes that generate ATP and downregulating anabolic processes that consume ATP but are not acutely needed for survival. SIRT1 in turn is best known historically for producing genetic changes that mediate the increase in longevity caused by calorie restriction. Although the two molecules have been studied intensively for many years, only recently has it become apparent that they have similar effects on diverse processes such as cellular fuel metabolism, inflammation, and mitochondrial function. In this review we will examine the evidence that these similarities occur because AMPK and SIRT1 both regulate each other and share many common target molecules. In addition, we will discuss the clinical relevance of these interactions and in particular the possibility that their dysregulation predisposes to disorders such as type 2 diabetes and atherosclerotic cardiovascular disease and is a target for their therapy.adenosine 5=-monophosphate-activated protein kinase; sirtuin 1; metabolic syndrome; mitochondrial function; insulin resistance; peroxisome proliferator-activated receptor-␥ coactivator-1␣; type 2 diabetes; atherosclerosis AMP-ACTIVATED PROTEIN KINASE (AMPK) is a fuel-sensing enzyme that is activated by decreases in a cell's energy state as reflected by an increased AMP/ATP ratio. When activated, it initiates metabolic and genetic events that restore ATP levels by stimulating processes that generate ATP (e.g., fatty acid oxidation) and inhibiting others that consume ATP but are not acutely required for survival (e.g., triglyceride and protein synthesis, cell proliferation) (40). In addition, AMPK sets in motion changes in mitochondrial biogenesis and function (37) that could more chronically increase the ability of a cell to generate ATP and diminish oxidative stress and other potentially adverse cellular events (78). The sirtuins are a family of evolutionarily conserved NAD ϩ -dependent histone/protein deacetylases that are also widely regarded as fuel-sensing molecules. They have many actions (see below) but are perhaps best known for their role in mediating the increase in longevity caused by caloric restriction in various species, including yeast, worms, and possibly mammals (21, 85). Seven sirtuins have been identified in mammalian cells. Of these the most studied and the focus of this review is silent information regulator T1 (SIRT1).AMPK and the sirtuins are present in all eukaryotic cells and probably have coexisted throughout evolution (7, 29). Although both molecules have been studied intensively, the similarities in their regulation and in their actions on such diverse processes as cellular metabolism, inflammation, and mitochondrial function have only recently become apparent. In this review we will examine the evidence that these similarities occur, ...

Resveratrol may protect against metabolic disease through activating SIRT1 deacetylase. Because we have recently defined AMPK activation as a key mechanism for the beneficial effects of polyphenols on hepatic lipid accumulation, hyperlipidemia, and atherosclerosis in type 1 diabetic mice, we hypothesize that polyphenol-activated SIRT1 acts upstream of AMPK signaling and hepatocellular lipid metabolism. Here we show that polyphenols, including resveratrol and the synthetic polyphenol S17834, increase SIRT1 deacetylase activity, LKB1 phosphorylation at Ser 428 , and AMPK activity. Polyphenols substantially prevent the impairment in phosphorylation of AMPK and its downstream target, ACC (acetyl-CoA carboxylase), elevation in expression of FAS (fatty acid synthase), and lipid accumulation in human HepG2 hepatocytes exposed to high glucose. These effects of polyphenols are largely abolished by pharmacological and genetic inhibition of SIRT1, suggesting that the stimulation of AMPK and lipid-lowering effect of polyphenols depend on SIRT1 activity. Furthermore, adenoviral overexpression of SIRT1 stimulates the basal AMPK signaling in HepG2 cells and in the mouse liver. AMPK activation by SIRT1 also protects against FAS induction and lipid accumulation caused by high glucose. Moreover, LKB1, but not CaMKK␤, is required for activation of AMPK by polyphenols and SIRT1. These findings suggest that SIRT1 functions as a novel upstream regulator for LKB1/AMPK signaling and plays an essential role in the regulation of hepatocyte lipid metabolism. Targeting SIRT1/LKB1/ AMPK signaling by polyphenols may have potential therapeutic implications for dyslipidemia and accelerated atherosclerosis in diabetes and age-related diseases. AMPK (AMP-activated protein kinase)2 serves as a sensor of cellular energy status, being activated by increased AMP/ATP ratio or by the upstream kinases, LKB1 (the tumor suppressor kinase), CaMKK␤ (Ca 2ϩ /calmodulin-dependent protein kinase kinase ␤), and TAK1 (transforming growth factor-␤-activated kinase-1) (1-7). Our previous studies demonstrated that dysfunction of hepatic AMPK induced by hyperglycemia represents a key mechanism for hepatic lipid accumulation and hyperlipidemia associated with diabetes (8, 9). Also, metformin, an antidiabetic drug, lowers systemic and hepatic lipids via activating LKB1/AMPK signaling (2, 8, 10). Our recent studies with human hepatocytes and type 1 diabetic LDL receptor-deficient (LDLR Ϫ/Ϫ ) mice have shown that polyphenols strongly stimulate hepatic AMPK and reduce lipid accumulation, which in turn attenuates hyperlipidemia and atherosclerosis in diabetic mice (9). Therefore, AMPK activation by polyphenols or metformin may be at least partially responsible for their therapeutic benefits on hyperlipidemia in diabetes (2,8,9). Resveratrol also stimulates AMPK in neurons (11). However, rapid activation of AMPK by polyphenols has been shown to be independent of altered adenine nucleotide levels (9, 11). Also, resveratrol activates AMPK in intact cells via an indirect mech...

SIRT1, a histone/protein deacetylase, and AMP-activated protein kinase (AMPK) are key enzymes responsible for longevity and energy homeostasis. We examined whether a mechanistic connection exists between these molecules that involves the major AMPK kinase LKB1. Initial studies demonstrated that LKB1 is acetylated in cultured (HEK293T) cells, mouse white adipose tissue, and rat liver. In the 293T cells, SIRT1 overexpression diminished lysine acetylation of LKB1 and concurrently increased its activity, cytoplasmic/nuclear ratio, and association with the LKB1 activator STRAD. In contrast, short hairpin RNA for SIRT1, where studied, had opposite effects on these parameters. Mass spectrometric analysis established that acetylation of LKB1 occurs on multiple, but specific, lysine residues; however, only mutation of lysine 48 to arginine, which mimics deacetylation, reproduced all of the effects of activated SIRT1. SIRT1 also affected downstream targets of LKB1. Thus its overexpression increased AMPK and acetyl-CoA carboxylase phosphorylation, and conversely, RNA interference-mediated SIRT1 knockdown reduced AMPK phosphorylation and that of another LKB1 target MARK1. Consistent with the results in cultured cells, total LKB1 lysine acetylation was decreased by 60% in the liver of 48-h starved rats compared with starved-refed rats, and this was associated with modest but significant increases in both LKB1 and AMPK activities. These results suggest that LKB1 deacetylation is regulated by SIRT1 and that this in turn influences its intracellular localization, association with STRAD, kinase activity, and ability to activate AMPK.LKB1 is a serine-threonine protein kinase that phosphorylates and activates 13 downstream kinases (1), one of which is AMP-activated protein kinase (AMPK), 2 a key enzyme that regulates cellular energy state, growth, inflammation, and mitochondrial function (2). LKB1, when not associated with other proteins, is located predominantly in the nucleus because of its N-terminal nuclear localization signal. However, LKB1 activation takes place predominantly in the cytoplasm, after it complexes with STRAD (STE-related adapter) and MO25 (mouse protein 25) (1, 3). Once activated, LKB1 has been demonstrated to phosphorylate AMPK on Thr-172, an event required for its activation (4). On the other hand, no specific mechanism for regulating the activation and inactivation of the kinase activity of LKB1 has been described. Indeed, it has been suggested that LKB1 may be constitutively active and that its effects on AMPK phosphorylation (e.g. in contracting muscle) may be governed by the action of phosphatases (1, 20). SIRT1, a class III NAD ϩ -dependent histone/protein deacetylase, has been implicated in the longevity induced by caloric restriction in species ranging from Caenorhabditis elegans to rodents (5). It has been suggested that it may work in part by activating AMPK (5). The expression and deacetylation activities of SIRT1 are enhanced by increases in NAD ϩ levels or the NAD ϩ /NADH ratio, such as occur dur...

The popular inverse probability weighting method in causal inference is often hampered by extreme propensity scores, resulting in biased estimates and excessive variance. A common remedy is to trim patients with extreme scores. However, such methods are often sensitive to the choice of cutoff points and discard a large proportion of the sample. The implications to bias and precision of the treatment effect estimate are unclear. These problems are mitigated by the newly developed overlap weighting method. Overlap weights emphasize the target population with the most overlap in observed characteristics between treatments, by continuously down-weighting the units in the tails of the propensity score distribution. We use simulations to compare overlap weights to standard inverse probability weighting with trimming, in terms of bias, variance and confidence interval coverage. A range of propensity score distributions are considered, including settings with substantial non-overlap and extreme values. To facilitate practical implementation, we further provide a consistent estimator for standard error of the treatment effect estimated using overlap weighting.