Category: AMPK



Clinical trials have shown that angiotensin II receptor blockers reduce the new onset of diabetes in hypertensives; however, the underlying mechanisms remain unknown. We investigated the effects of telmisartan on peroxisome proliferator activated receptor γ (PPAR-δ) and the adenosine monophosphate (AMP)-activated protein kinase (AMPK) pathway in cultured myotubes, as well as on the running endurance of wild-type and PPAR-δ-deficient mice. Administration of telmisartan up-regulated levels of PPAR-δ and phospho-AMPKα in cultured myotubes. However, PPAR-δ gene deficiency completely abolished the telmisartan effect on phospho-AMPKαin vitro. Chronic administration of telmisartan remarkably prevented weight gain, enhanced running endurance and post-exercise oxygen consumption, and increased slow-twitch skeletal muscle fibres in wild-type mice, but these effects were absent in PPAR-δ-deficient mice. The mechanism is involved in PPAR-δ-mediated stimulation of the AMPK pathway. Compared to the control mice, phospho-AMPKα level in skeletal muscle was up-regulated in mice treated with telmisartan. In contrast, phospho-AMPKα expression in skeletal muscle was unchanged in PPAR-δ-deficient mice treated with telmisartan. These findings highlight the ability of telmisartan to improve skeletal muscle function, and they implicate PPAR-δ as a potential therapeutic target for the prevention of type 2 diabetes.

© 2011 The Authors Journal of Cellular and Molecular Medicine © 2011 Foundation for Cellular and Molecular Medicine/Blackwell Publishing Ltd.

[PubMed – indexed for MEDLINE]



The World Antidoping Agency (WADA) has introduced some changes in the 2012 prohibited list. Among the leading innovations to the rules are that both 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (peroxisome proliferator-activated receptor-δ [PPAR-δ]-5′ adenosine monophosphate-activated protein kinase [AMPK] agonist) and GW1516 (PPAR-δ-agonist) are no longer categorized as gene doping substances in the new 2012 prohibited list but as metabolic modulators in the class “Hormone and metabolic modulators.” This may also be valid for the angotensin II receptor blocker telmisartan. It has recently been shown that telmisartan might induce similar biochemical, biological, and metabolic changes (e.g., mitochondrial biogenesis and changes in skeletal muscle fiber type) as those reported for the former call of substances. We suspect that metabolic modulators abuse such as telmisartan might become a tangible threat in sports and should be thereby targeted as an important antidoping issue. The 2012 WADA prohibited list does not provide telmisartan for a potential doping drug, but arguments supporting the consideration to include them among “metabolic modulators” are at hand.

[PubMed – in process]


Telmisartan shows antihypertensive and several pleiotropic effects that interact with metabolic pathways. In the present study we tested the hypothesis that telmisartan prevents adipogenesis in vitro and weight gain in vivo through activation of peroxisome proliferator-activated receptor (PPAR)-delta-dependent pathways in several tissues. In vitro, telmisartan significantly upregulated PPAR-delta expression in 3T3-L1 preadipocytes in a time- and dose-dependent manner. Other than enhancing PPAR-delta expression by 68.2+/-17.3% and PPAR-delta activity by 102.0+/-9.0%, telmisartan also upregulated PPAR-gamma expression, whereas neither candesartan nor losartan affected PPAR-delta expression. In vivo, long-term administration of telmisartan significantly reduced visceral fat and prevented high-fat diet-induced obesity in wild-type mice and hypertensive rats but not in PPAR-delta knockout mice. Administration of telmisartan did not influence food intake in mice. Telmisartan influenced several lipolytic and energy uncoupling related proteins (UCPs) and enhanced phosphorylated protein kinase A and hormone sensitive lipase but reduced perilipin expression and finally inhibited adipogenesis in 3T3-L1 preadipocytes. Telmisartan-associated reduction of adipogenesis in preadipocytes was significantly blocked after PPAR-delta gene knockout. Chronic telmisartan treatment upregulated the expressions of protein kinase A, hormone-sensitive lipase, and uncoupling protein 1 but reduced perilipin expression in adipose tissue and increased uncoupling protein 2 and 3 expression in skeletal muscle in wild-type mice but not in PPAR-delta knockout mice. We conclude that telmisartan prevents adipogenesis and weight gain through activation of PPAR-delta-dependent lipolytic pathways and energy uncoupling in several tissues.

Arctigenin Efficiently Enhanced Sedentary Mice Treadmill Endurance

Physical inactivity is considered as one of the potential risk factors for the development of type 2 diabetes and other metabolic diseases, while endurance exercise training could enhance fat oxidation that is associated with insulin sensitivity improvement in obesity. AMP-activated protein kinase (AMPK) as an energy sensor plays pivotal roles in the regulation of energy homeostasis, and its activation could improve glucose uptake, promote mitochondrial biogenesis and increase glycolysis. Recent research has even suggested that AMPK activation contributed to endurance enhancement without exercise. Here we report that the natural product arctigenin from the traditional herb Arctium lappa L. (Compositae) strongly increased AMPK phosphorylation and subsequently up-regulated its downstream pathway in both H9C2 and C2C12 cells. It was discovered that arctigenin phosphorylated AMPK via calmodulin-dependent protein kinase kinase (CaMKK) and serine/threonine kinase 11(LKB1)-dependent pathways. Mice treadmill based in vivo assay further indicated that administration of arctigenin improved efficiently mice endurance as reflected by the increased fatigue time and distance, and potently enhanced mitochondrial biogenesis and fatty acid oxidation (FAO) related genes expression in muscle tissues. Our results thus suggested that arctigenin might be used as a potential lead compound for the discovery of the agents with mimic exercise training effects to treat metabolic diseases.
Physical inactivity is considered as one of the risk factors for the development of type 2 diabetes and other metabolic diseases. It is known that endurance exercise training could lead to fiber type transformation, mitochondrial biogenesis, angiogenesis and other adaptive changes in skeletal muscle [1], [2], thus further enhancing fat oxidation that is associated with improvement of insulin sensitivity in obesity [3]. Currently, at least 60% of the global population fails to achieve the daily minimum recommendation of 30 min moderate intensity of physical activity, and within these people the risk of getting the related chronic diseases including type 2 diabetes increases by 1.5 times [4]. Therefore, it has become valuable to discover active agents that would mimic the effects of exercise training to prevent or treat metabolic diseases.
AMP-activated protein kinase (AMPK) is a heterotrimeric serine/threonine protein kinase with three subunits (α, β, γ) [5]. As the major molecular sensor for AMP/ATP ratio in cells, AMPK plays a pivotal role in the regulation of energy metabolism [6]. AMPK activation switches on ATP-producing processes (such as glucose uptake, mitochondrial biogenesis and glycolysis) and inhibits ATP-consuming anabolic processes (such as protein synthesis and sterol synthesis) [7]. AMPK phosphorylation is regulated by a series of upstream AMPK kinases, including serine/threonine kinase (LKB1) [8], Tak1 kinase and two calmodulin-dependent protein kinase kinases (CaMKKα and CaMKKβ) [9], [10].
Recently, it was reported that AMPK activation could improve mice endurance in the absence of exercise training [11]. Under endurance training condition, skeletal muscle suffers a number of changes, such as glucose consumption decreasing, main energy source transition from glucose to fatty acid utilization, mitochondrial biogenesis increasing and fiber-type switch [11], [12]. AMPK activation could increase peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) gene expression or the direct phosphorylation of PGC-1α in skeletal muscle [6], [13]. PGC-1α is known to be involved in the regulation of mitochondrial biogenesis, respiration, hepatic gluconeogenesis and other biological processes by interaction with several transcription factors, such as ERRα, NRF1, NRF2 and PPARs [14], [15]. PGC-1α null mice showed muscle dysfunction, abnormal weight control and hepatic steatosis [16], [17]. Although skeletal muscle was not an active lipogenic organ as liver, exercise training or AMPK activation has been also reported to promote fatty acid synthesis and oxidation as evidenced by up-regulation of varied main enzymes such as pyruvate dehydrogenase kinase 4 (PDK4), stearoyl-CoA desaturase-1 (SCD-1), fatty acid synthetase (FAS) and muscle carnitine palmitoyltransferase I (mCPT1b) within related pathways in skeletal muscles [18][20]. Oxidation of fatty acid exports more energy than metabolism of glucose, and alteration of utilizing this energy substrate could contribute to exercise tolerance [11], [21]. Additionally, AMPK activation changes skeletal muscle myofiber type composition, which mimics the fiber type switch induced by endurance training [11], [22]. Therefore, all the above findings suggested that AMPK might act as a key mediator of endurance training-induced changes, which was also confirmed further by the result that treatment of AMPK agonist AICAR at a dose of 500 mg/kg/day could induce metabolic genes expression and enhance running endurance [11].
Arctigenin (ATG, Fig. 1A) is a phenylpropanoid dibenzylbutyrolactone lignan extracted from the traditional herb Arctium lappa L. (Compositae) with anti-cancer and anti-inflammatory effects [23][26]. As a new type of antitumor agent, arctigenin could block the unfolded protein response (UPR) by reducing the expression of UPR-related genes, such as C/EBP homologous protein (CHOP), activating transcription factor 4 (ATF4) and glucose-regulated protein of 78 kDa (GRP78) under glucose deprivation [23][25]. Its anti-inflammatory effect mainly acts by inhibiting type I-IV allergic inflammation and pro-inflammatory enzymes in vitro and in vivo [26].
Figure 1
Figure 1

Arctigenin (ATG) increased AMPK phosphorylation in H9C2 and C2C12 cells.
In the present study, we discovered that arctigenin could increase AMPK phosphorylation and up-regulate its downstream-pathway related genes mRNA levels to promote mitochondrial biogenesis and fatty acid synthesis and oxidation in vitro and in vivo, subsequently leading to the mice treadmill endurance enhancement. Cell based assay revealed that arctigenin increased AMPK phosphorylation targeting the CaMKK and LKB1-dependent pathways. Our results thus demonstrated that arctigenin might be used as a lead compound for the discovery of the agents with mimic exercise training effects to treat metabolic diseases.
Arctigenin increased AMPK phosphorylation in H9C2 and C2C12 cells
As reported, AMPK activation switched on ATP-producing processes to enhance endurance [7], [11]. With this information, we thus constructed the phospho-AMPK activator screening platform (Text S1), based on which our lab in-house natural product library was screened out and the natural product arctigenin (Fig. 1A) was finally identified to efficiently activate AMPK phosphorylation (Fig. S1). As shown in Fig. 1B and C, arctigenin dose-dependently increased AMPK phosphorylation in both H9C2 and C2C12 muscle cells, while had no effects on total AMPK.
Arctigenin activated PGC-1α transcription via up-regulating AMPK phosphorylation
PGC-1α was known to act as the master regulator in mitochondrial biogenesis and skeletal muscle adaptation [27][29]. Since AMPK activation by actual exercise or pharmacological treatment could lead to up-regulation of PGC-1α gene expression [30], [31], we investigated the potential effects of arctigenin on PGC-1α mRNA level in both H9C2 and C2C12 cells. Compared with DMSO-treated group, arctigenin incubation could dose-dependently up-regulate PGC-1α mRNA levels in both cell lines (Fig. 2A). By considering that AMPK phosphorylated PGC-1α directly at theronine-177 and serine-538, which are required for induction of PGC-1α promoter [6], we thus examined whether arctigenin could increase PGC-1α transcription through regulating its promoter activity by luciferase assay. As shown in Fig. S2A, the relative PGC-1α promoter activity was highly increased in arctigenin treated group compared with the DMSO group in HEK293T cells.
Figure 2
Figure 2

Arctigenin (ATG) increased PGC-1α transcription via enhancing AMPK phosphrylation.
PGC-1α as a key regulator in energy metabolism could respond to varied physical stimuli, such as muscle contraction, cold stress and overfeeding [32][35]. It could be regulated by AMPK, p38 MAPK or NF-κB pathway [11], [36][38]. Since arctigenin has been determined to increase AMPK phosphorylation and PGC-1α transcription, we thus wondered whether the effect of arctigenin on PGC-1α was dependent on its effect on AMPK. Therefore, we examined the effects of arctigenin on PGC-1α mRNA level and PGC-1α promoter activity together with AMPK inhibitor compound C incubation in the related cells. As shown in Fig. 2B–E and Fig. S2B, treatment of compound C almost inhibited the arctigenin-induced AMPK phosphorylation and completely blocked the arctigenin-induced up-regulation of PGC-1α mRNA level and promoter activity, implying that the effect of arctigenin on PGC-1α transcription regulation was dependent on its role in AMPK phosphorylation.
These results thereby indicated that arctigenin could induce PGC-1α transcription in skeletal muscle and cardiac muscle cell lines via up-regulating AMPK phosphorylation.
Arctigenin increased mitochondrial biogenesis and fatty acid oxidation genes expression
Since arctigenin has been determined to induce PGC-1α transcription via up-regulating AMPK phosphorylation, this result thereby implied that arctigenin might play a potential role in promoting mitochondrial biogenesis and function. To further evaluate this hypothesis, the mRNA levels of the related genes were detected in both H9C2 and C2C12 cell lines with results listed in Fig. 3A and B. Estrogen-related receptor α (ERRα), a nuclear receptor activated by interacting with PGC-1α, was reported to control the expression of nuclear genes encoding mitochondrial proteins (NUGEMPs) [39], [40]. We found ERRα mRNA level was obviously elevated by arctigenin treatment. At the same time, as a typical NUGEMP and significant component of electron transport chain (ETC) in mitochondrial, cytochrome c mRNA was also induced by arctigenin administration. Furthermore, as also indicated in Fig. 3A, arctigenin significantly activated the mRNA levels of PDK4, SCD1, FAS and mCPT1b which are key enzymes in fatty acid synthesis and oxidation for promotion of energy source transforming from glucose to fatty acid in C2C12 cells at 40 µM, while the regulation of arctigenin in these genes was not effectively in H9C2 cells for lack of significant difference in SCD1 and FAS mRNA levels between high dose of arctigenin administration group (40 µM) and DMSO control (Fig. 3B).
Figure 3
Figure 3

Arctigenin (ATG) promoted mitochondrial biogenesis and FAO related gene expression.
To clarify whether arctigenin regulated mitochondrial biogenesis and FAO related genes mRNA levels through its effect on AMPK phosphorylation, we examined the influence of arctigenin on these genes together with AMPK inhibitor (compound C) incubation in C2C12 and H9C2 cells. As indicated in Fig. S3 and S4, treatment of compound C almost inhibited the arctigenin-induced genes expression except mCPT1b in H9C2 and SCD1 in C2C12 cells. These results thus demonstrated that arctigenin increased mitochondrial biogenesis and fatty acid oxidation genes mRNA levels mainly through its effect on AMPK phosphorylation.
Therefore, all our results suggested that arctigenin could activate PGC-1α transcription and increase mitochondrial biogenesis and fatty acid oxidation genes expression in both skeletal muscle and cardiac muscle cells.
Arctigenin regulated the related genes expression not via affecting transcriptional activity of PPARδ
As reported, AMPK catalytic subunit over-expression evidently promoted basal and ligand-dependent transcription of PPARδ [11], indicative of that PPARδ might be also involved in AMPK related gene expression. The fact that AICAR (AMPK activator) synergistically increased mice endurance and gene expression with GW501516 (PPARδ agonist) further implied the potential of AMPK-PPARδ signaling axis. Here, we also examined whether arctigenin could induce the related PPARδ involved gene expression with mammalian one-hybrid and transcriptional activation assay systems. As shown in Fig. S5, arctigenin failed to regulate the co-activator recruitment or transcriptional activity of exogenous or endogenous PPARδ.
Therefore, all above-mentioned results suggested that arctigenin regulating the relative gene expression is related to the enhancement of AMPK phosphorylation, while might not to the promotion of PPARδ transcriptional activity.
Arctigenin enhanced AMPK phosphorylation through CaMKK and LKB1-dependent pathways
Considering that arctigenin could activate AMPK phosphorylation, we subsequently investigated the potential regulation of this natural product against the relevant AMPK-involved pathways. As determined, directly activating AMPK in an allosteric manner and/or indirectly promoting Thr172 phosphorylation of AMPK both contributed to the AMPK activation. Bear that in mind, we examined the recombinant AMPK enzyme activity with or without arctigenin incubation in vitro to investigate whether arctigenin could influence the conformation of AMPK thus inducing AMPK activity. As indicated in Fig. 4A, arctigenin had no effect on the recombinant AMPK enzyme activity, while AMPK agonist A-769662 as a positive control significantly increased the phosphorylation of its substrate SAMS [41], [42], implying that arctigenin was not an AMPK ligand and indirectly activated AMPK.
Figure 4
Figure 4

Arctigenin (ATG) enhanced AMPK phosphorylation through CaMKK and LKB1-dependent pathway.
As reported, AMPK has an obligate requirement for phosphorylation by an upstream kinase on Thr-172 in the α-subunit catalytic domain [5]. Since the above assay has indicated that arctigenin activated AMPK in an indirect manner, we further explored the potential signaling responsible for arctigenin-induced AMPK activation. Regarding the fact that LKB1 and CaMKK are within the main identified kinases in AMPK upstream pathway and play pivotal roles in regulation of AMPK phosphorylation [43], we thus addressed these two kinases related assays. Firstly, we determined whether CaMKK activation was necessary for arctigenin-induced AMPK phosphorylation. As shown in Fig. 4B, pretreatment of HEK293T, a model cell usually used in mechanism studies, with the selective CaMKK inhibitor STO-609 [9] obviously attenuated the arctigenin-induced AMPK phosphorylation. To investigate whether LKB1 pathway might participate in the arctigenin-induced AMPK activation, we carried out LKB1 knock-down assay in HEK293T cells as reported [44]. The results in Fig. 4C revealed that LKB1 knock-down in HEK293T cells efficiently down-regulated the arctigenin-induced AMPK phosphorylation.
Taken together, our results thereby suggested that arctigenin stimulated AMPK phosphorylation via CaMKK and LKB1-dependent pathways.
Arctigenin efficiently enhanced sedentary mice treadmill endurance
AMPK was reported as an “exercise mimetic”, whose pharmacological activator-AICAR was ever tested to provide several benefits of exercise in sedentary mice [11]. Since arctigenin has been determined able to effectively activate AMPK phosphorylation, we thereby evaluated whether this natural product could enhance mice endurance. To address this issue, we performed the treadmill exhaustion test among the selected mice (All mice were examined regarding the treadmill performance before arctigenin administration, and those mice whose running time was too long or too short compared with the average were eliminated to reduce the potential effects by the inherent variation) after 6-week arctigenin administration (8 mg/kg). Running time and distance till fatigue were estimated as maximal endurance capacity. As shown in Fig. 5A and B, arctigenin administration brought on approximately an increase of 40% in mean fatigue time and 65% in mean fatigue distance, further indicating that arctigenin efficiently enhanced sedentary mice treadmill endurance.
Figure 5
Figure 5

Arctigenin (ATG) elevated mice treadmill performance without inducing myofiber type conversion in skeletal muscle.
During the treatment, mice showed similar basal behavior, food consumption and body weight (Fig. S6A and B). In addition, to investigate the preliminary toxicity of arctigenin, TNFα and IL-6 levels in mice serum were tested. The results revealed that there was no difference in those two inflammatory factors comparing arctigenin with vehicle groups (Fig. S6C and D). At the same time, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were also detected. As shown in Fig. S6E and F, no change was found for AST while ALT elevated with arctigenin administration, suggestive of the potential hepatotoxicity of arctigenin in ip administration. Therefore, these results demonstrated that arctigenin elevated sedentary mice endurance efficiently while no significant toxicity was observed.
Arctigenin could not induce skeletal muscle fiber-type changes
As reported, endurance exercise, AMPK mutation with persistent activation and PGC-1α over-expression in transgenic mice driven by muscle creatine kinase (MCK) promoter could induce myofiber type switch [12], [22], [45]. We thereby considered that arctigenin might change myofiber type construction in skeletal muscle tissues. However, we failed to find any obvious conversions in the mRNA levels of four different myosin heavy chain (MHC) isoforms (MHCI as a marker to represent type I myofibers, MHCIIa, MHCIIx and MHCIIb as markers to represent type II myofibers) in gastrocnemius (Fig. 5C) and quadriceps (Fig. 5D) comparing arctigenin administration group with vehicle group. Additionally, no obvious changes were found in myofibers compositions between arctigenin treatment group and vehicle group in ATPase staining (Fig. S7). These results thus suggested that arctigenin could not affect myofiber type proportion.
Arctigenin induced AMPK phosphorylation, mitochondrial biogenesis and FAO pathway in vivo
To further investigate the potential regulative mechanism of arctigenin regarding its improvement of mice treadmill endurance, we addressed the relevant tissue-based assays. Compared with vehicle group, arctigenin (8 mg/kg) administration enhanced AMPK phosphorylation in gastrocnemius (Fig. 6A), quadriceps (Fig. 6B) and cardiac muscles (Fig. 6C). The results shown in Fig. 6D–F suggested that arctigenin up-regulated the mRNA levels of PGC-1α and ERRα and the protein levels of cytochrome c in gastrocnemius, quadriceps and cardiac muscles. Additionally, mRNA levels of SCD1, PDK4 and mCPT1b were also determined to be obviously elevated in gastrocnemius and quadriceps muscles with arctigenin treatment, consistent with the cell based results. Notably, there was also a tendency for higher average FAS mRNA levels in arctigenin treated group but without significance. To further identify the regulation of arctigenin on FAO pathway, the protein level of uncoupling protein 3 (UCP3) that is also involved in FAO induction [46] was examined. As shown in Fig. 6A–C, UCP3 expression was obviously elevated in gastrocnemius and quadriceps muscles.
Figure 6
Figure 6

Arctigenin (ATG) enhanced AMPK phosphorylation, mitochondrial biogenesis and FAO pathway in vivo.
As fatty acid synthesis and storage related genes were up-regulated, we thereby hypothesized that arctigenin treated mice might contain more fatty acid in skeletal muscle for instant oxidation to supply energy source during exercise. Bearing that in mind, we thus detected fatty acid levels in skeletal muscles (gastrocnemius and quadriceps) and found that arctigenin treatment could enhance fatty acid storage in gastrocnemius evidently while in quadriceps without significance (Fig. S8A and B).
Therefore, all those in vitro and in vivo results thereby suggested that arctigenin could enhance AMPK phosphorylation, mitochondrial biogenesis and FAO pathway related genes expression, finally leading to the enhancement of exercise-free endurance.
Over the past decades, aerobic endurance exercise has been potently highlighted concerning its significance in the clinical amelioration of many disease symptoms, such as glucose metabolism in type 2 diabetes [47], dyslipidemia in atherosclerosis [48] and hypertension in stroke, acute myocardial infarction or cardiac insufficiency [49], [50]. Nevertheless, the inability to afford definite intensity of physical exercise has been always the obstacle to make profits of exercises [51]. Discovery of active agents that would mimic the reprogramming metabolism induced by exercise training is one of the effective strategies to overcome these obstacles.
Actual exercises could result in activation of kinases/phoshatases signaling pathway, nuclear translocation of transcription/translation factors, up-regulation of NUGEMPs and mtDNA-encoded proteins, and augmentation of muscle aerobic capacity [52]. AMPK is activated during physical activities to promote down-stream metabolic reprogramming (e.g. mitochondria biogenesis, fatty acid oxidation promotion and myofiber type switch), and considered as a significant mediator in skeletal muscle adaptations [6], [7], [22]. Therefore, pharmacological activation of AMPK may provide benefits of endurance exercise without actual physical activities, which was ever confirmed by endurance enhancing activity of its activators AICAR [11]. It was found that AMPK activator-AICAR, which is metabolized to an AMP mimetic in cell, could increase sedentary mice treadmill endurance at a high dosage of 500 mg/kg/day via intraperitoneal injection and up-regulate genes linked to oxidation metabolism via AMPK-PPARδ signaling axis [11]. Resveratrol, a natural polyphenolic product derived from grapes, could also increase mice aerobic capacity without exercise at a dose of 400 mg/kg/day orally [53] targeting SIRT1 and subsequently regulating its downstream pathway, which was reported to tightly couple with AMPK [54].
Compared with synthetic compounds, small molecules from natural sources are featured by their large-scale of structure diversity [55]. Therefore, we performed the screening of the efficient phosphor-AMPK activator targeting our in-house natural product library and finally discovered that arctigenin dose-dependently increased AMPK phosphorylation in vitro (Fig. 1B and C) and in vivo (Fig. 6A–C).
Arctigenin was extracted from Arctium lappa L., which has been widely used in traditional Chinese medicine [56]. Previous studies have illustrated that arctigenin was active in anti-viral infection, anti-tumor, anti-inflammation, and neuroprotection [23][26], [57][59]. Here we reported that arctigenin promoted AMPK phosphorylation on Thr172 site through CaMKK and LKB1-dependent pathways (Fig. 4). Recently, AMPK Ser485/491 phosphorylation was also reported involving in regulation of AMPK activity [60], [61]. To clarify whether arctigenin could affect AMPK phosphorylation on Ser485/491, we examined Ser485/491 phosphorylation levels of AMPK in arctigenin treated cells. As shown in Fig. S9, arctigenin did not change AMPK Ser485/491 phosphorylation in HEK293T, H9C2 or C2C12 cell lines, implying that arctigenin activated AMPK phosphorylation on Thr172 without impacting AMPK Ser485/491 phosphorylation.
We also found that arctigenin significantly enhanced sedentary mice running endurance at a dose of 8 mg/kg/day (Fig. 5A and B). Furthermore, cytochrome c protein and mRNA levels were typically increased in mitochondrial biogenesis, and FAO related genes (SCD1, PDK4, FAS and mCPT1b) mRNA levels were also obviously elevated in muscle tissues (Fig. 6). Fatigue induced by treadmill running was primarily developed from periphery tissues and featured by rapid clearance of intracellular ATP and relative insufficiency of oxidation metabolism in cardiovascular and skeletal muscle system [62]. We thus concluded that promotion of mitochondrial biogenesis and FAO linked gene expression induced by arctigenin attributed to mice prolonged running time and distance.
Myofiber type switch induced by endurance training was ever considered as one of the reasons for exercise tolerance [12], [22], [45], but we could not find any differences of fiber type composition in skeletal muscle tissues (gastrocnemius and quadriceps) between arctigenin-treatment and vehicle groups (Fig. 5C–D and Fig. S5). Compared with AMPK persistent activating mutation that led to myofiber type switch in vivo [45], the indirect phosphorylation of AMPK by arctigenin might only share part of the down-stream genes response with AMPK persistent activating mutation, which resulted in unchangeable myofiber type composition in arctigenin administrated group. Meanwhile, PGC-1α was also reported playing a pivotal role in myofiber type transformation via activating calcineurin signaling pathway [22]. Although arctigenin obviously up-regulated PGC-1α transcription in vitro (Fig. 2A and B) and in vivo (Fig. 6D–F), it might not mobilize calcium/calcineurin pathway or totally activate PGC-1α interaction with related proteins for promotion of the myofiber type switch. It is noted that arctigenin up-regulated mitochondrial biogenesis related genes (such as ERRα and cytochrome c) both in cardiac and skeletal muscle tissues, but elevated FAO related genes only in skeletal muscle tissues, which might be tentatively attributed to the tissue specificity responding to arctigenin in fatty acid metabolism.
In summary, we demonstrated that arctigenin could efficiently increase rodent sedentary treadmill endurance via enhancing AMPK phosphorylation. This natural product induced the accommodation of mitochondrial biogenesis and FAO pathway to promote mitochondrial oxidative capacity without actual physical activities as summarized in Fig. 7. Our results have provided additional understanding of pharmacological functions for arctigenin and traditional Chinese medicine Arctium lappa L., and suggested the potential of arctigenin as a lead compound for anti-chronic metabolic disease (e.g. obesity or type 2 diabetes) drug discovery.
Figure 7
Figure 7

A proposed model demonstrating arctigenin (ATG)-induced endurance enhancement mechanism.
Materials and Methods
Ethics Statement
All animal experiments were carried out in accordance with the Regulations of Experiments Animal Administration issued by the State Committee of Science and Technology of the People’s Republic of China. Permit numbers: SCXK (HU) 2008-0017; SYXK (HU) 2008-0049. This study was approved by Science and Technology Commission of Shanghai Municipality.
Restriction enzymes were purchased from New England Biolabs. Cell culture plastic ware was purchased from Corning Inc. DMEM, fetal bovine and horse serums were purchased from Invitrogen. Compound C and STO609 were obtained from Sigma. Calcium Phosphate Cell Transfection Kit was obtained from Beyotime. RNAiso, RT reagent Kit and SYRB Premix Ex Taq were purchased from TaKaRa. Dual Luciferase Assay System was obtained from Promega. Anti-cytochrome c, anti-phospho-AMPK (Thr172), anti-AMPKα1/α2, and anti-LKB1 antibodies were purchased from Cell Signaling Technology. Anti-CaMKK antibody was purchased from Senta Cruz Biotechnology. HEK293T, H9C2 and C2C12 cells were obtained from ATCC.
Cell culture and differentiation
As a typical cardiac muscle cell line, H9C2 was derived from embryonic BD1X rat heart tissue and not differentiated in our study. C2C12 was a subclone of the mouse myoblast cell line and differentiated in DMEM with 2% horse serum, forming contractile myotubes and expressing characteristic muscle proteins. Differentiated myotubes were used in our experiments.
H9C2 and C2C12 cell lines were maintained in DMEM supplemented with 10% fetal bovine serum and the cells were grown at 37°C in an environment of 5% CO2. To induce myoblast fusion and myotubes differentiation, C2C12 myoblasts were switched to differentiation medium when 100% confluent in 6-well plate. Differentiation medium was exchanged every 2 days for 6 days before experimental manipulation.
Western Blot analysis
Tissues were lysed with lysis buffer containing 25 mmol/L Tris-HCl (PH 7.5), 150 mmol/L NaCl, 1 mmol/L Na3VO4, 1% Triton X-100 and a protease inhibitor cocktail (Sigma-Aldrich). Protein concentrations were determined using a BCA protein assay kit (Pierce, Rockford, IL). Equal amounts of lysates or supernatants of the whole cell extracts were fractionated by SDS-PAGE and transferred to Hybond-c nitrocellulose membrane (Amersham Bioscience). The membranes were blocked for one hour at room temperature and then incubated overnight at 4°C in TBST buffer (5% milk) containing related antibody. The membranes were then incubated for an hour at room temperature in TBST buffer (5% milk) containing anti-rabbit IgG or anti-mouse IgG (Jackson-ImmunoResearch, West Grove, PA). Blots were visualized by incubation with SuperSignal West Dura chemiluminescence kit (Pierce Biotechnology) and exposing to light-sensitive film.
The bands were quantified as “intensity×area” using Image-Pro Plis software (MediaCybernetics) and statistically analyzed. SE was calculated from three repeats of the experiments or five individuals of arctigenin treated and vehicle groups.
RT-PCR and quantitative real-time PCR
Total RNA was extracted from administrated cells or mice tissues using RNAiso (TaKaRa) reagent in accordance with the Kit instruction. cDNA was synthesized by RT reagent Kit (TaKaRa) and real-time PCR was performed using SYRB Premix Ex Taq (TaKaRa) on DNA Engene Opticon TM2 system (MJ Research, Waltham, MA, USA). The primers were listed as follows:
PGC-1α: sense 5′- gcccggtacagtgagtgttc-3′, anti 5′- ctgggccgtttagtcttcct-3′.
ERRα: sense 5′- ctcagctctctacccaaacgc-3′, anti 5′- ccgcttggtgatctcacactc-3′.
Cytochrome c: sense 5′- cagcttccattgcggacac-3′, anti 5′- ggcactcacggcagaatgaa-3′.
PDK4: sense 5′- agggaggtcgagctgttctc-3′, anti 5′- ggagtgttcactaagcggtca-3′.
FAS: sense 5′- ggaggtggtgatagccggtat-3′, anti 5′- tgggtaatccatagagcccag-3′.
mCPT1b: sense 5′- tgggactggtcgattgcatc-3′, anti 5′- tcagggtttgtcggaagagaga-3′.
SCD1: sense 5′- ttcttgcgatacactctggtgc-3′, anti 5′- cgggattgaatgttcttgtcgt-3′.
GAPDH: sense 5′- acagcaacagggtggtggac-3′, anti 5′- tttgagggtgcagcgaactt-3.
MHCI: sense 5′- ccttggcaccaatgtcccggctc-3′, anti 5′- gaagcgcaatgcagatgcggtg-3.
MHCIIa: sense 5′- atgagctccgacgccgag-3′, anti 5′- tctgttagcatgaactggtaggcg-3.
MHCIIx: sense 5′- aaggagcaggacaccagcgccca-3′, anti 5′- atctctttggtcactttcctgct-3.
MHCIIb: sense 5′- gtgatttctcctgtcacctctc-3′, anti 5′- ggaggaccgcaagaacgtgctga-3.
Luciferase assay
For evaluation of the effects of arctigenin on PGC-1α promoter, HEK293T cells (24-well plate) were transfected with pGL3-PGC-1α-luc together with PRL-SV40 and refreshed with normal medium 5 hours later. After transfection, HEK293T cells were incubated with indicated concentration of arctigenin for 24 h. Luciferase activity was measured using Dual Luciferase Assay kit.
AMPK enzymatic activity assay
The recombinant AMPK activity was assayed using the modified conventional approach in non-radioactive way (such as Jun N-terminal kinase 2 (Jnk2α2), casein kinase 1, Protein kinase A (PKA), etc.) [63], as outlined in Fig. S10.
In the assay, the substrate SAMS was used according to the literature methods [41], [42]. The recombinant AMPK isoform α2β1γ1 was purchased from Invitrogen. AMPK was diluted to 400 ng/ml in assay buffer (pH 7.4, containing 15 mM HEPES, 20 mM NaCl, 1 mM EGTA, 0.02% Tween) and pre-incubated with arctigenin (0.2, 2, 20 µM) or A-769662 (1 µM) for 30 min on ice. The kinase reaction was initiated by adding ATP (50 µM) and SAMS (50 µM) at room temperature for 30 min. The produced ADP reflecting the AMPK enzyme activity was thus measured by the ADP Hunter Plus Assay kit (DiscoverX), and the fluorescent signal was detected with an M5 Multi-Detection Reader using excitation and emission wavelengths of 530 and 590 nm.
Animal experiment
C57BL/6J male mice at 6 weeks of age were purchased from Shanghai Experimental Animal Center, Chinese Academy of Sciences, and acclimated to SPF microisollators for 2 days before any experimental intervention. Mice were accommodated under standard conditions (strict 12:12-h light-dark cycle, 22°C, 60% humidity) in plastic cages and provided with water and food ad libitum.
All 30 mice were adapted to treadmill running for 10 min at 10 m/min at week -1 (5 days/week) avoiding unexpected accidents and the first fatigue test was conducted at week 0. For fatigue test, mice ran at 10 m/min for 5 min and 15 m/min for 10 min. After the initial warm-up period, exercise intensity was increased by 5 m/min every 30 min from 20 m/min until mice could not be prompted to continue running by moderate electric stimulation (less than 0.1 milliampere) and stayed at electrode for at least 10 sec. After the first fatigue test, 20 mice with moderate endurance capacity were selected from total 30 mice and divided into arctigenin administration and vehicle treatment groups (n = 10/group). Before final fatigue assay, sedentary mice were treated with arctigenin (8 mg/kg, body weight/day) or vehicle (sterilized 0.9% Sodium Chloride containing 5% Tween-80) daily via intraperitoneal injection for 6 weeks. Mice endurance capacity was estimated by treadmill (Litai Science and Technology Inc. Exer6, Hangzhou, China) running time and distance until fatigue [11], [62]. After 6-week arctigenin administration, the last fatigue test was performed according to the same protocol as before (n = 8/group). For investigation of the relevant gene changes in tissues, two mice in each group as control were chosen escaping the fatigue test.
Tissue collection
Animals were euthanized 72 h after the last bout of exercise. Gastrocnemius, quadriceps and heart muscles were thus isolated, frozen, and stored at −80°C until further analysis.
Statistical analysis
All data were reported as mean ± standard deviation of the mean (SD). Data were analyzed in either one-way ANOVA with an appropriate post hoc test for comparison of multiple groups or unpaired student’s t test for comparison of two groups as described in figure legends (Graphpad Prism software).
Supporting Information
Figure S1

Arctigenin (ATG) enhanced AMPK phosphorylation in HEK293T cells. HEK293T cells were incubated with indicated concentrations of arctigenin (0-40 µM) for 30 min, phospho- and total AMPK were then detected by western blotting. The results shown are representative of three independent experiments. The bands were quantified using Image-Pro Plus software. Values are means ± SE. **, p<0.05; ***, p<0.005; one-way ANOVA.
Figure S2

Arctigenin (ATG) activated PGC-1α transcription via up-regulating AMPK phosphorylation. A. When the confluence reached 30∼40% (24-well plate), HEK293T cells were transiently transfected with pGL3-PGC-1α promoter-Luc and SV40. 5 hours later, cells were refreshed with medium supplemented with arctigenin (1, 10, 40 µM) or DMSO and incubated for 24 hours before Luciferase assays as described in “Materials and methods”. B. After transfection, HEK293T cells were administrated with or without 20 µM compound C for 1 hour before and during the incubation with actigenin (40 µM) for 24 hours before Luciferase assays as described in “Materials and methods”. **, p<0.01. ##, p<0.01: for compound C and arctigenin co-incubation group versus arctigenin treated group; student’s t test.
Figure S3

Effects of arctigenin (ATG) on ERRα, cytochrome c, PDK4, SCD1, FAS and mCPT1b were subjective to AMPK phosphorylation in H9C2. H9C2 cells were treated with or without 20 µM compound C for 1 hour before and during the incubation with actigenin (20 µM) for 24 hours. After harvested, mRNA levels of ERRα (A), cytochrome c (B), SCD1 (C), PDK4 (D), FAS (E) and mCPT1b (F) were analyzed. The results shown are representative of three independent experiments. Values are means ± SD. *, p<0.05. #, p<0.05: for compound C and arctigenin co-incubation group versus arctigenin treated group; student’s t test.
Figure S4

Effects of arctigenin (ATG) on ERRα, cytochrome c, PDK4, SCD1, FAS and mCPT1b were subjective to AMPK phosphorylation in C2C12. Differentiated C2C12 cells were administrated with or without 20 µM compound C for 1 hour before and during the incubation with actigenin (20 µM) for 24 hours. After harvested, mRNA levels of ERRα (A), cytochrome c (B), SCD1 (C), PDK4 (D), FAS (E) and mCPT1b (F) were analyzed. The results shown are representative of three independent experiments. Values are means ± SD. *, p<0.05; **, p<0.01; ***, p<0.005. #, p<0.05; ##, p<0.01; ###, p<0.005: for compound C and arctigenin co-incubation group versus arctigenin treated group; student’s t test.
Figure S5

Arctigenin (ATG) failed to regulate the co-activator recruitment and transcriptional activity of PPARδ. A. HEK293T cells were transfected with UAS-TK-Luc, pCMX-Gal4DBD-PPARδ-LBD and pRL-SV40 followed by treatment of DMSO, GW501516 (PPARδ agonist), and varied concentrations of arctigenin for 24 hours. B. HEK293T cells were transfected with pAdTrack-PPARδ, pcDNA3.1-RXRα, pSV-PPRE-Luc and pRL-SV40 and then incubated with DMSO, GW501516 (PPARδ agonist), and varied concentrations of arctigenin for 24 hours. C. HEK293T cells were transfected with pSV-PPRE-Luc and pRL-SV40, and incubated with DMSO, GW501516 (PPARd agonist), and varied concentrations of arctigenin for 24 hours. Relative luciferase activities were measured as described in Text S1. The results shown are representative of three independent experiments. Values are means ± SD. *, p<0.05; **, p<0.01; ***, p<0.005; one-way ANOVA.
Figure S6

Effects of arctigenin (ATG) on diet, weight, inflammation and liver toxicity of mice. A. Daily food intake of each group was analyzed (n = 10/group). B. Weight change in each group was measured (n = 10/group). C. D. Serum from mice in each group was collected and levels of TNFα (C) and IL-6 (D) were analyzed (n = 10/group). E. F. Activities of ALT (E) and AST (F) were measured (n = 10/group). Values are means ± SE. *, p<0.5; **, p<0.01; student’s t test.
Figure S7

Arctigenin (ATG) failed to induce skeletal muscle fiber-type change. Metachromatical staining of frozen cross-sections from gastrocnemius and quadriceps in vehicle and arctigenin treated groups. The results shown are representative of three independent experiments. Dark-brown stained type I fibers were indicated by arrows.
Figure S8

Arctigenin (ATG) enhanced fatty acid storage in gastrocnemius. Free fatty acid in gastrocnemius (A) or quadriceps (B) of each group was analyzed (n = 7/group). Values are means ± SD. *, p<0.5; student’s t test.
Figure S9

Arctigenin (ATG) failed to impact the phosphorylation of AMPK on Ser485/491 sites. HEK293T, H9C2 and differentiated C2C12 cells were incubated with indicated concentrations of arctigenin (0–40 µM) for 30 min, AMPK (Thr172), AMPK (Ser485/491) and total AMPK were then detected by western blotting. The results shown are representative of three independent experiments.

Figure S10

A scheme demonstrating recombinant AMPK activity assay approach.


The increased energy required for acute moderate exercise by skeletal muscle (SkM) is derived equally from enhanced fatty acid (FA) oxidation and glucose oxidation. Availability of FA also influences contracting SkM metabolic responses. Whole body glucose turnover and SkM glucose metabolic responses were determined in paired dog studies during 1) a 30-min moderate exercise (maximal oxygen consumption of ∼60%) test vs. a 60-min low-dose 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (AICAR) infusion, 2) a 150-min AICAR infusion vs. modest elevation of FA induced by a 150-min combined intralipid-heparin (IL/hep) infusion, and 3) an acute exercise test performed with vs. without IL/hep. The exercise responses differed from those observed with AICAR: plasma FA and glycerol rose sharply with exercise, whereas FA fell and glycerol was unchanged with AICAR; glucose turnover and glycolytic flux doubled with exercise but rose only by 50% with AICAR; SkM glucose-6-phosphate rose and glycogen content decreased with exercise, whereas no changes occurred with AICAR. The metabolic responses to AICAR vs. IL/hep differed: glycolytic flux was stimulated by AICAR but suppressed by IL/hep, and no changes in glucose turnover occurred with IL/hep. Glucose turnover responses to exercise were similar in the IL/hep and non-IL/hep, but SkM lactate and glycogen concentrations rose with IL/hep vs. that shown with exercise alone. In conclusion, the metabolic responses to acute exercise are not mimicked by a single dose of AICAR or altered by short-term enhancement of fatty acid supply.

Exercise vs. AICAR Infusion

With 30 min of moderate exercise, plasma glucose, insulin [4.1 ± 1.0 (before) vs. 3.0 ± 0.5 mU/l (after)], and glucagon [75 ± 17 (before) vs. 93 ± 21 ng/l (after); P = not significant (NS)] were unchanged; however, plasma FFA, lactate, and glycerol all rose significantly (P > 0.01) (Fig. 1). In contrast, with the 60-min AICAR infusion, plasma glucose fell (P = 0.05), insulin and glucagon were unchanged [insulin = 4.7 ± 0.5 (before) vs. 6.1 ± 0.8 mU/l (after), P = NS; glucagon = 53 ± 11 (before) vs. 68 ± 11 ng/l (after), P = NS], plasma FFA decreased significantly (P < 0.01), with no change in glycerol, and lactate rose sharply (P < 0.01) (Fig. 1). Thus the incremental responses of FFA and glycerol to exercise and AICAR were significantly different and in the opposite directions (Fig. 1). There was a greater than twofold increase (P < 0.02–0.01) in HGP, Rd, MCRg, and GFexog after the 30-min moderate exercise test (Fig. 2). In contrast, although these parameters also increased after the 60-min AICAR infusion (P < 0.02), the responses were considerably less (∼30–40%) and significantly (P < 0.01) smaller for the AICAR than for the exercise responses (Fig. 2). The metabolic responses to AICAR at 150 min were similar to those obtained at 60 min, with only a minor attenuation of MCRg by 150 min. SkM substrate concentrations of intracellular glucose and lactate did not change with exercise alone, whereas G6P significantly increased (change of 1.0 ± 0.3 mmol/l; P = 0.05) and glycogen decreased (change of –35 ± 36 mmol/kg dry wt; P = 0.06). In contrast, during AICAR infusion, a decrease in intracellular SkM glucose (change of –0.20 ± 0.0.9 mmol/l; P = 0.06) was noted, but no significant changes occurred in G6P, lactate, or glycogen. However, the absolute decrease of the glycogen concentration was significantly different for exercise compared with that for AICAR (change of –35 ± 36 vs. 9 ± 36 mmol/kg dry wt; P = 0.06). There were no changes in AMPK-α1 and AMPK-α2 activities with exercise or AICAR infusion (data not shown), but ACCβ-pSer221 was elevated by ∼35% [from basal level of 1.05 ± 0.02 to 1.38 ± 0.15 (P ≤ 0.05) with exercise and to 1.32 ± 0.15 (arbitrary units) (P = 0.06) with AICAR].

View larger version (17K):
In this window
In a new window

Fig. 1. Paired comparisons of absolute plasma substrate responses and incremental changes (Δ) after either 30-min moderate exercise [exer; ∼60% maximal oxygen consumption (O2max); open bars] or 60-min low-dose 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (AICAR) infusion (hatched bars) in 6 dogs. A: plasma (P) glucose. B: plasma free fatty acid (FFA). C: lactate. D: glycerol. *P = 0.05; **P < 0.01 vs. the within-group basal state; ††P < 0.01 between groups.

View larger version (19K):
In this window
In a new window

Fig. 2. Paired comparisons of absolute responses and incremental changes (Δ) of hepatic glucose production (HGP; A), metabolic clearance rate of glucose (MCRg; B), Rd (C), and glycolytic flux (GFexog; D) to 30-min moderate exercise (∼60% O2max) (open bars) or 60-min AICAR infusion (hatched bars) in 6 dogs. *P < 0.02; **P < 0.01 vs. the within-group basal state; ††P < 0.01 between groups.

IL/Hep Infusion vs. Saline and AICAR Infusions

The 150-min IL/hep infusion alone raised plasma FFA by ∼60% and glycerol by ∼80% (P < 0.05–0.01) (Table 1). This resulted in a small but significant fall in fasting glucose (P ≤ 0.05), with no changes in insulin or glucagon (Table 1). There were also no changes induced by the IL/hep infusion in HGP, Rd, and MCRg, but GFexog fell significantly (P < 0.05) by ∼20% (Table 1). The effects of increasing FA supply compared with the AICAR infusion at 150 min were different (Table 1). Plasma glucose did not change; however, in the IL/hep group, the change in plasma lactate was markedly reduced (P < 0.001) and, as expected, FFA and glycerol were raised, compared with the responses observed in the AICAR studies (P ≤ 0.05–0.01) (Table 1). The modest significant increments observed for HGP, Rd, MCRg, and GFexog with AICAR were not observed in the IL/hep studies (Table 1). There was no relationship between the changes in plasma glucagon and HGP with the combined AICAR and IL/hep groups (r = 0.40, P = 0.22). After the IL/hep infusion, there were no changes (Δ) in intramuscular substrates (Δglucose = –0.05 ± 0.27 mmol/l, ΔG6P = 0.7 ± 0.7 mmol/kg dry wt, Δlactate = 0.51 ± 0.50 mmol/l, Δglycogen = 15 ± 33 mmol/kg dry wt, and ΔACCβ-pSer221 = 0.03 ± 0.05 arbitrary units).

Table 1. Comparison of the absolute and incremental effects of 60-min and 150-min AICAR infusion and 150-min saline or IL/heparin infusion on plasma substrates and hormones, glucose turnover, and skeletal muscle substrates in 6 resting normal dogs

Exercise With or Without IL/Hep Infusion

The exercise-induced responses of plasma substrates in the presence or absence of excess FA supply were similar for plasma glucose, glycerol, lactate, insulin, and glucagon. No exercise-induced increments for plasma FA were observed in the IL/hep group, since the FFA levels were already raised preexercise above those seen postexercise for the non-IL/hep group (Fig. 3). On the background of the lipid-induced lower basal GFexog (Table 1), the glucose turnover responses to exercise were similar for the IL/hep and non-IL/hep groups for HGP, Rd, MCRg, and GFexog (Fig. 4). Only small responses in the SkM substrates with exercise performed with and without IL/hep infusion were noted [Δglucose = 0.3 ± 0.1 vs. 0.0 ± 0.2 mmol/l, ΔG6P = 0.8 ± 0.6 vs. 1.0 ± 0.3 mmol/l, Δlactate = 2.6 ± 1.1 vs. 0.0 ± 0.3 mmol/l (P = 0.07), Δglycogen = 30 ± 37 vs. –35 ± 36 mmol/kg dry wt (P = 0.09), and ΔACCβ-pSer221 = 0.29 ± 0.06 vs. 0.23 ± 0.15 arbitrary units, respectively].

View larger version (19K):
In this window
In a new window

Fig. 3. Paired comparisons of absolute responses and incremental changes of plasma substrates to 30-min moderate exercise (∼60% O2max), either in the absence [open bars, with saline (Sal)] or presence (hatched bars) of intralipid-heparin (IL/hep) infusion in 6 dogs. A: plasma glucose. B: plasma FFA. C: plasma lactate. D: plasma glycerol. *P < 0.05; **P < 0.01 vs. the within-group basal state; †P ≤ 0.06 between groups.

View larger version (21K):
In this window
In a new window

Fig. 4. Paired comparisons of absolute responses and incremental changes of HGP (A), MCRg (B), Rd (C), and GFexog (D) to moderate exercise (∼60% O2max), either in the absence (open bars) or presence (hatched bars) of IL/hep infusion in 6 dogs. **P < 0.01 vs. the within-group basal state; †P ≤ 0.05 vs. the corresponding basal study without IL/hep.


AICAR BLOWOUT SALE! Contact us for purchase options

     100 mg: $45

     1000 mg: $350

     3 grams: $800

shipping: $20 flat rate worldwide

AICAR, aminoimidazole carboxamide ribonucleotide, acts as an agonist to AMP-activated protein kinase; AMP-activated protein kinase, also known as AMPK, is an enzyme with an important role in cellular homeostasis and energy regulation.[1]  AMPK acts through a variety of means to ultimately stimulate liver fatty oxidation, ketogenesis, beta-cell modulation of insulin secretion, and other functions within the body.  AICAR has been shown to stimulate glucose uptake and reduce apoptosis by reducing reactive oxygen compounds within cells.[2][3]

In a breakthrough study in 2008, Narkar et al of the Salk Institute discovered that AICAR significantly improves the performance of mice in endurance-type exercise by converting fast-twitch muscle fibers to the more energy-efficient, fat-burning, slow-twitch type. They also found that AICAR and GW1516, when given to “sedentary” mice, activated 40% of the genes that were turned on when mice were given GW1516 and made to exercise. As a result a publicity storm about “exercise pills” and “exercise in a pill” ensued.  The World Anti-Doping Agency now lists both compounds on their prohibited list (since 2009), and the lead researcher of the breakthrough study cooperated in providing data to make possible a urinalysis test to detect AICAR.[4][5]

Figure 5 Likely effects of AICAR (and other AMPK activators) on glucose homoeostasis in vivo

AICAR (1) stimulates glucose uptake into muscle through the membrane recruitment of Glut4, (2) inhibits hepatic glucose output and triacylglycerol synthesis, (3) inhibits both glucose uptake and lipolysis by adipose tissue, (4) acutely suppresses insulin release from pancreatic islets, and (5) activates glucose-responsive neurons in the paraventricular and arcuate nuclei of the hypothalamus, potentially stimulating appetite. Effects (1)–(4) probably explain the glucose-lowering effects of AICAR (and contribute to the effects of metformin and glitazones) and are likely to be beneficial in Type II diabetes. Effects (4) and (5) may be contra-indicated. FA, fatty acids.

AICAR has a number of other experimental/clinical and research chemical uses as it is expressed in a variety of tissue types.  Bai et al found that “data demonstrate that AICAR-initiated AMPK activation may represent a promising alternative to our current approaches to suppressing intestinal inflammation in IBD.”[6]

Guo et al found “results suggest[ing] a mechanism by which AICAR inhibits the proliferation of EGFRvIII expressing glioblastomas and point toward a potential therapeutic strategy for targeting EGFR-activated cancers.”[7]

An original study by Pold et al offers additional hope that AICAR could offer important treatment potential for humans:

Five-week-old, pre-diabetic ZDF rats underwent daily treadmill running or AICAR treatment over an 8-week period and were compared with an untreated group. In contrast to the untreated, both the exercised and AICAR-treated rats did not develop hyperglycemia during the intervention period. Whole-body insulin sensitivity, as assessed by a hyperinsulinemic-euglycemic clamp at the end of the intervention period, was markedly increased in the exercised and AICAR-treated animals compared with the untreated ZDF rats (P < 0.01). In addition, pancreatic beta-cell morphology was almost normal in the exercised and AICAR-treated animals, indicating that chronic AMPK activation in vivo might preserve beta-cell function. Our results suggest that activation of AMPK may represent a therapeutic approach to improve insulin action and prevent a decrease in beta-cell function associated with type 2 diabetes.[8]


[1]Corton JM, Gillespie JG, Hawley SA, Hardie DG. “5-aminoimidazole-4-carboxamide ribonucleoside. A specific method for activating AMP-activated protein kinase in intact cells?”. Eur. J. Biochem. 229 (2): 558–65.  1995.

[2]Lemieux K, Konrad D, Klip A, Marette A. “The AMP-activated protein kinase activator AICAR does not induce GLUT4 translocation to transverse tubules but stimulates glucose uptake and p38 mitogen-activated protein kinases alpha and beta in skeletal muscle”. Faseb J. 17 (12): 1658–65. 2003.

[3]Kim JE, Kim YW, Lee IK, Kim JY, Kang YJ, Park SY.  “AMP-activated protein kinase activation by

5-aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside (AICAR) inhibits palmitate-induced endothelial cell apoptosis through reactive oxygen species suppression”. J. Pharmacol. Sci. 106 (3): 394–403. 2008.

[4]Narkar VA, Downes M, Yu RT, Embler E, Wang YX, Banayo E, Mihaylova MM, Nelson MC, Zou Y, Juguilon H, Kang H, Shaw RJ, Evans RM. “AMPK and PPARdelta agonists are exercise mimetics”. Cell 134 (3): 405–15. 2008. [5]WADA 2009 Prohibited List: WADA PROHIBITED LIST PDF (PDF Document).

[6]Bai A, Yong M, Ma Y, Ma A, Weiss C, Guan Q, Bernstein C, Peng Z. Novel Anti-Inflammatory Action of 5-Aminoimidazole-4-carboxamide ribonucleoside with protective effect in DSS-induced acute and chronic colitis. J Pharmacol Exp Ther. 2010 Mar 17.

[7]Guo D, Hildebrandt IJ, Prins RM, Soto H, Mazzotta MM, Dang J, Czernin J, Shyy JY, Watson AD, Phelps M, Radu CG, Cloughesy TF, Mischel PS.  The AMPK agonist AICAR inhibits the growth of EGFRvIII-expressing glioblastomas by inhibiting lipogenesis.Proc Natl Acad Sci U S A. 2009 Aug 4;106(31):12932-7.

[8]Pold R, Jensen LS, Jessen N, Buhl ES, Schmitz O, Flyvbjerg A, Fujii N, Goodyear LJ, Gotfredsen CF, Brand CL, Lund S. Long-term AICAR administration and exercise prevents diabetes in ZDF rats.  Diabetes. 2005 Apr;54(4):928-34.

*The latter article is intended for educational / informational purposes only. THIS PRODUCT IS INTENDED AS A RESEARCH CHEMICAL ONLY. This designation allows the use of research chemicals strictly for in vitro testing and laboratory experimentation only. Bodily introduction of any kind into humans or animals is strictly forbidden by law.

Role of AMPK2 in basal, training-, and AICAR-induced GLUT4, hexokinase II, and mitochondrial protein expression in mouse muscle

Am J Physiol Endocrinol Metab 292: E331–E339, 2007. First published September 5, 2006;

We investigated the role of AMPK2 in basal, exercise training-, and AICAR-induced protein expression of GLUT4, hexokinase II (HKII), mitochondrial markers, and AMPK subunits. This was conducted in red (RG) and white gastrocnemius (WG) muscle from wild-type (WT) and 2-knockout (KO) mice after 28 days of activity wheel running or daily AICAR injection. Additional experiments were conducted to measure acute activation of AMPK by exercise and AICAR. At basal, mitochondrial markers were reduced by 20% in 2-KO muscles compared with WT. In both muscle types, AMPK2 activity was increased in response to both stimuli, whereas AMPK1
activity was increased only in response to exercise. Furthermore, AMPK signaling was estimated to be 60–70% lower in 2-KO compared with WT muscles. In WG, AICAR treatment increased HKII, GLUT4, cytochrome
c, COX-1, and CS, and the 2-KO abolished the AICARinduced increases, whereas no AICAR responses were observed in RG. Exercise training increased GLUT4, HKII, COX-1, CS, and HAD protein in WG, but the 2-KO did not affect training-induced increases. Furthermore, AMPK1, -2, -1, -2, and -3 subunits were reduced in RG, but not in WG, by 30–60% in response to exercise training. In conclusion, the 2-KO was associated with an 20% reduction in mitochondrial markers in both muscle types and abolished AICAR-induced increases in protein expression in WG. However, the 2-KO did not reduce traininginduced increases in HKII, GLUT4, COX-1, HAD, or CS protein in WG, suggesting that AMPK2 may not be essential for metabolic adaptations of skeletal muscles to exercise training.

5-adenosine monophosphate-activated protein kinase-2; glucose transporter-4; 5-aminoimidazole-1--D-ribofuranoside; exercise training; mitochondrial proteins; skeletal muscle

BIOCHEMICAL ADAPTATIONS of skeletal muscle to regular physical activity include an increase in mitochondrial oxidative enzyme capacity and an increase in key proteins involved in glucose uptake, such as glucose transporter-4 (GLUT4) and hexokinase II (HKII) (41, 51). The intracellular pathways involved in eliciting these exercise training-induced increases in protein expression and mitochondrial content remain largely unknown. The 5-AMPactivated protein kinase (AMPK) has been proposed as a signaling molecule involved in transmitting an “exercise signal” to the nuclei of the muscle cell (15, 49). This is in part based on the finding that AMPK is activated by in vitro electrical induced contractions of rodent skeletal muscle (8, 44) and by in vivo exercise in human (11, 50) and rodent skeletal muscle (48).
Chronic activation of AMPK by the adenosine analog 5-aminoimidazole- 4-carboxamide-1--D-ribofuranoside (AICAR) or the creatine analog -guanadinopropionic acid (-GPA) in resting rat and mouse muscles increases transcription of metabolic genes and expression of metabolic enzymes as well as mitochondrial density mimicking effects of exercise training (4, 15, 20, 49, 52). The observation that pharmacologically induced upregulation of peroxisome proliferator-activated receptor-  coactivator-1 (PGC-1) and -aminolevulinate synthase mRNA, cytochrome c (cyt c) protein, and mitochondrial density is abolished in mouse muscle overexpressing a kinasedead AMPK construct suggests a causal role of AMPK in these responses (52). The AMPK-dependent increase in PGC-1 is of particular interest because this nuclear transcription modulator is shown to be important in coordinating muscular adaptations in lipid oxidation and mitochondrial function to exercise training (23, 24) and to be involved in regulating GLUT4 protein expression (28, 30).
Several studies indicate that AMPK increases content of target proteins by increasing transcriptional activity of their respective genes. For instance, arterial infusion of AICAR in rodent hindlimb muscle stimulates HKII and uncoupling protein- 3 (UCP3) gene transcription (43), and an acute injection of AICAR increases muscle GLUT4 mRNA in rodents (5). Furthermore, a chronic treatment of rat with -GPA treatment is followed by an increase in nuclear respiratory factor-1 (NRF-1) DNA binding activity correlated by an increase in expression of NRF-1 targets and mitochondrial density in muscle (4).
NRF-1 is activated by binding with PGC-1, which in turn leads to increased transcription of nuclear genes encoding subunits of the mitochondrial respiratory chain and components of the mitochondrial transcription and replication machinery (for review, see Refs. 22 and 42).
Altogether, it seems evident that activation of AMPK in resting muscle specifically increases the protein content of several metabolic enzymes as well as mitochondrial density.
Therefore, the purpose of the present study was to investigate the importance of the catalytic AMPK2 subunit in expression of HKII and GLUT4 protein, as well as mitochondrial enzymes in skeletal muscle in the basal state, and after exercise training and chronic AICAR treatment. This was investigated in AMPK2 whole body knockout (2- KO) and corresponding wild-type (WT) mice that had undertaken a 28-day program of activity wheel exercise training or daily AICAR injections.


Metabolic disorders, including type 2 diabetes and obesity, represent major health risks in industrialized countries. AMP-activated protein kinase (AMPK) has become the focus of a great deal of attention as a novel therapeutic target for the treatment of metabolic syndromes, because AMPK has been demonstrated to mediate, at least in part, the effects of a number of physiological and pharmacological factors that exert beneficial effects on these disorders. Thus, the identification of a compound that activates the AMPK pathway would contribute significantly to the treatment and management of such syndromes. In service of this goal, we have screened a variety of naturally occurring compounds and have identified one compound, cryptotanshinone, as a novel AMPK pathway activator. Cryptotanshinone was originally isolated from the dried roots of Salvia militorrhiza, an herb that is used extensively in Asian medicine and that is known to exert beneficial effects on the circulatory system. For the first time, in the present study, we have described the potent antidiabetic and antiobesity effects of cryptotanshinone, both in vitro and in vivo. Our findings suggest that the activation of the AMPK pathway might contribute to the development of novel therapeutic approaches for the treatment of metabolic disorders such as type 2 diabetes and obesity.



Internal Medicine, UNICAMP.



Metformin is a widely-used antidiabetic drug whose anti-cancer effects, mediated by the activation of AMPK and reduction of mTOR signaling, have become noteworthy. Chemotherapy produces genotoxic stress and induces p53 activity, which can cross-talk with AMPK/mTOR pathway. Herein, we investigate whether the combination of metformin and paclitaxel has an effect in cancer cell lines.


Human tumors were xenografted into SCID mice and the cancer cell lines were treated with only paclitaxel or metformin, or a combination of both drugs. Western Blotting, flow cytometry and immunohistochemistry were then used to characterize the effects of the different treatments.


The results presented herein, demonstrate that the addition of metformin to paclitaxel leads to quantitative potentialization of molecular signaling through AMPK and a subsequent potent inhibition of the mTOR signaling pathway. Treatment with metformin and paclitaxel resulted in an increase in the number of cells arrested in the G2/M phase of the cell cycle, decreased the tumor growth and increased apoptosis in tumor-bearing mice, when compared to individual drug treatments.


We have provided evidence for a convergence of metformin and paclitaxel induced signaling at the level of AMPK. This mechanism illustrates how different drugs may cooperate to augment anti-growth signals, and suggests that target activation of AMPK by metformin may be a compelling ally in cancer treatment.

Curcumin activates AMPK and suppresses gluconeogenic gene expression in hepatoma cells.
Kim T, Davis J, Zhang AJ, He X, Mathews ST.

Department of Nutrition and Food Science, Boshell Diabetes and Metabolic Diseases, Research Program, Auburn University, Auburn, AL 36849, USA.

Curcumin, the bioactive component of curry spice turmeric, and its related structures possess potent anti-oxidant and anti-inflammatory properties. Several lines of evidence suggest that curcumin may play a beneficial role in animal models of diabetes, both by lowering blood glucose levels and by ameliorating the long-term complications of diabetes. However, current understanding of the mechanism of curcumin action is rudimentary and is limited to its anti-oxidant and anti-inflammatory effects. In this study we examine potential anti-diabetic mechanisms of curcumin, curcumin C3 complex), and tetrahydrocurcuminoids (THC). Curcuminoids did not exert a direct effect on receptor tyrosine kinase activity, 2-deoxy glucose uptake in L6-GLUT4myc cells, or intestinal glucose metabolism measured by DPP4/alpha-glucosidase inhibitory activity. We demonstrate that curcuminoids effectively suppressed dexamethasone-induced phosphoenol pyruvate carboxy kinase (PEPCK) and glucose6-phosphatase (G6Pase) in H4IIE rat hepatoma and Hep3B human hepatoma cells. Furthermore, curcuminoids increased the phosphorylation of AMP-activated protein kinase (AMPK) and its downstream target acetyl-CoA carboxylase (ACC) in H4IIE and Hep3B cells with 400 times (curcumin) to 100,000 times (THC) the potency of metformin. These results suggest that AMPK mediated suppression of hepatic gluconeogenesis may be a potential mechanism mediating glucose-lowering effects of curcuminoids.

[PubMed – indexed for MEDLINE]

Safety and anti-inflammatory activity of curcumin: a component of tumeric (Curcuma longa).
Chainani-Wu N.

Department of Stomatology, University of California, San Francisco, CA 94143-0658, USA.

Tumeric is a spice that comes from the root Curcuma longa, a member of the ginger family, Zingaberaceae. In Ayurveda (Indian traditional medicine), tumeric has been used for its medicinal properties for various indications and through different routes of administration, including topically, orally, and by inhalation. Curcuminoids are components of tumeric, which include mainly curcumin (diferuloyl methane), demethoxycurcumin, and bisdemethoxycurcmin.

The goal of this systematic review of the literature was to summarize the literature on the safety and anti-inflammatory activity of curcumin.

A search of the computerized database MEDLINE (1966 to January 2002), a manual search of bibliographies of papers identified through MEDLINE, and an Internet search using multiple search engines for references on this topic was conducted. The PDR for Herbal Medicines, and four textbooks on herbal medicine and their bibliographies were also searched.

A large number of studies on curcumin were identified. These included studies on the antioxidant, anti-inflammatory, antiviral, and antifungal properties of curcuminoids. Studies on the toxicity and anti-inflammatory properties of curcumin have included in vitro, animal, and human studies. A phase 1 human trial with 25 subjects using up to 8000 mg of curcumin per day for 3 months found no toxicity from curcumin. Five other human trials using 1125-2500 mg of curcumin per day have also found it to be safe. These human studies have found some evidence of anti-inflammatory activity of curcumin. The laboratory studies have identified a number of different molecules involved in inflammation that are inhibited by curcumin including phospholipase, lipooxygenase, cyclooxygenase 2, leukotrienes, thromboxane, prostaglandins, nitric oxide, collagenase, elastase, hyaluronidase, monocyte chemoattractant protein-1 (MCP-1), interferon-inducible protein, tumor necrosis factor (TNF), and interleukin-12 (IL-12).

Curcumin has been demonstrated to be safe in six human trials and has demonstrated anti-inflammatory activity. It may exert its anti-inflammatory activity by inhibition of a number of different molecules that play a role in inflammation.

[PubMed – indexed for MEDLINE]