American Diabetes Association – 72nd Scientific Sessions

June 8-12, 2012; Philadelphia, PA; Report – Novel Drug Development and Basic Science – Draft

Executive Highlights

Beyond SGLT-2 inhibitors, there were relatively few abstracts on clinically-staged novel therapies for the treatment of type 2 diabetes. Still, we were excited to hear new data presented for several candidates in development. In particular, 48-week results were reported from Stage 2 of the TINSAL-T2D study, which evaluated the efficacy and safety of the non-steroidal anti-inflammatory drug (NSAID) salsalate. Although some anti-inflammatory effects were conferred, 48 weeks of treatment resulted in only a modest reduction in A1c (0.24% beyond placebo), a trend toward increased SBP, significantly increased total cholesterol and LDL, significantly increased urinary albumin, an increased risk for mild hypoglycemia, and a modest increase in weight (~2 lbs) – overall, we don’t believe this bodes well for the therapy, especially given the weight gain. Separately, Metabolic Solutions Development Company (MSDC) reported promising 12-week phase 2b data for its lead PPAR-sparing insulin sensitizer MSDC- 0160 (966-P). Overall, the data demonstrated similar improvements in A1c and FPG with MSDC-0160 as with high-dose pioglitazone, but with less hemodilution and less weight gain. MSDC also revealed for the first time the mitochondrial target of its insulin sensitizers – a mitochondrial pyruvate carrier system that contains the proteins MPC1 and MPC2 (1096-P). Additionally, Eli Lilly highlighted additional clinical results for its glucagon receptor agonist LY2409021 (981-P). In a 12-week phase 2 study (n=87), treatment with LY2409021 led to robust reductions in A1c with a low risk for hypoglycemia and no changes in average weight, blood pressure, lipid levels, or triglyceride levels. However, similar to what was reported in a trial for LY2409021 at ADA 2011, dose-dependent increases in hepatic transaminases (elevations in these enzymes may be an indicator of liver damage) were observed. While transaminase levels returned to baseline after four weeks of washout and no indications of liver injury were detected, we wonder how clinically relevant this side effect will become during longer periods of treatment. Similar increases in hepatic transaminases were also observed with Merck’s former glucagon receptor agonist MK-0893 (also reported at ADA 2011), suggesting that the effect may be class related. Separately, we heard intriguing phase 1/2 results for a cord-blood derived multi-potent stem cell therapy (the Stem Cell Educator), which provided significant reductions in A1c, improvements in insulin resistance, and improvements in beta cell function four weeks following a single treatment (287-OR). Finally, additional data was presented from a multiple ascending dose study for Advinus’ liver selective glucokinase activator GKM-001, which showed dose-dependent and significant reductions in FPG and PPG with no hypoglycemia or changes in liver transaminases or triglycerides.

While basic science is not a major focus of ours at ADA, there were a few presentations of particular interest at this year’s meeting. One of our favorite presentations over the course of the conference was the Banting Lecture delivered by Dr. Bruce Spiegelman (Harvard Medical School, Boston, MA). Dr. Spiegelman gave a captivating lecture on the work conducted by his lab to elucidate the biology of brown fat regulation (including the discovery of irisin) and to develop brown-fat based therapeutics for the treatment of metabolic disease. Clearly passionate and tirelessly driven, we found listening to Dr. Spiegelman highly inspirational, and the lecture served as yet another reminder of how fortunate we are for the brilliant researchers and clinicians throughout the world working to advance diabetes and obesity care. We also enjoyed listening to the Outstanding Scientific Achievement Award Lecture given by Dr. David Altshuler (Broad Institute, Boston, MA). In front of a packed audience, Dr. Altshuler detailed his work using genome-wide association studies (GWAS) to identify novel genetic variants commonly associated with type 2 diabetes. Of greatest note, he asserted that attempting to use these genetic variants for prediction and prevention could be futile given the complex genetic basis for type 2 diabetes. Instead, he advocated for identifying loss-of-function mutations that confer protection against diseases to identify new drug targets, given that drugs more commonly inhibit rather than activate. Other basic science topics we found interesting included the roles of mitochondrial dysfunction, epigenetics, and inflammation in insulin resistance, the effects of insulin in the brain, and the molecular mechanisms underlying the beneficial effects of exercise.


Table of Contents 


Novel Drug Development

Oral Session: Novel Agents for Diabetes Management


Yong Zhao, MD, PhD (University of Illinois at Chicago, Chicago, IL)

Dr. Zhao presented the results from an open-label phase 1/2 study that examined the safety and efficacy of the Stem Cell Educator therapy in people with type 2 diabetes (n=25). The Stem Cell Educator circulates a patient’s blood through a closed loop device that separates lymphocytes from whole blood, co-cultures the lymphocytes with human cord blood-derived multipotent stem cells (which are believed to modulate immune responses), and returns the lymphocytes to the patient. Dr. Zhao indicated that the procedure duration was approximately eight hours. At baseline, average A1c was 8.5%, age was 50 years, and duration of diabetes was nine years. Following a single treatment with the Stem Cell Educator, A1c was statistically significantly reduced at week four (-0.6%; p=0.022) and at week 12 (- 1.4%; p<0.0001). Furthermore, more than 80% of participants treated with the Stem Cell Educator achieved an A1c <7.0% at week 12. Measures of both insulin resistance (HOMA-IR) and beta cell function (HOMA-B) were statistically significantly improved at week 12, as was AUC glucose following an OGTT. Although detailed safety data were not provided, Dr. Zhao indicated that the therapy was found to be very safe overall. We note that positive results from a phase 1/2 study in people with type 1 diabetes were published for the Stem Cell Educator therapy in January (Zhao et al., BMC Medicine 2012). Overall, this study demonstrated that single administration of the therapy could significantly improve A1c and C-peptide levels in both individuals with residual beta cell function and without residual beta cell function.

Questions and Answers

Q: Is there any introduction or transfer of the stem cells to patients?

A: No, there is no transfer. Both flow cytometry results and examinations of the device following treatment have suggested that the stem cells remain adherent to the bottom of the device.



Rashmi Barbhaiya, PhD (Advinus Therapeutics, Bangalore, India)

Dr. Barbhaiya reported results from a multiple ascending dose study for Advinus Therapeutics liver selective glucokinase activator GKM-001. As a reminder, this 14-day study randomized 60 people with type 2 diabetes to receive GKM-001 (ranging from 25 mg to 1000 mg) or placebo. At baseline, mean A1c was 9.0%, FPG was 176 mg/dl, and BMI was 26 kg/m2. Topline data reported in December showed that GKM-001 was effective at lower glucose levels across all doses tested without any incidence of hypoglycemia or other clinically relevant adverse events (for more details, see the December 15, 2011 Closer Look at In his presentation, Dr. Barbhaiya highlighted that treatment with GKM-001 provided: 1) dose dependent and significantly greater reductions in FPG, glucose excursions to a mixed-meal tolerance test, and 24-hours glucose over placebo. More specifically, the reductions in mean plasma glucose achieved were 9% with the 25 mg dose, 20% with the 1000 mg dose, and no change with placebo. Furthermore, Dr. Barbhaiya indicated that C-peptide responses to mixed-meal and oral glucose tolerance tests remained constant throughout the study in all GKM-001 treated arms, suggesting no activation of pancreatic glucokinase and induction of insulin secretion with the candidate therapy. On safety, Dr. Barbhaiya stated that no hypoglycemia, changes in liver transaminases, or changes in triglycerides were observed at any dose of GKM-001. Finally, regarding pharmacokinetics, he revealed that GKM-001 had a long half-life (~ 21 hours), that it was eliminated through both feces (75%) and the urine (25%), and that there were no food effects. Dr. Barbhaiya concluded by noting that a phase 2b study was currently in preparation, that Advinus would make a go/no go decision on global phase 3 development following the completion of phase 2 development in India, and that Advinus would seek a development and commercialization partner in the US and EU if GKM-001 progressed into phase 3 development.

Questions and Answers

Q: Did you measure body weight?

A: There was no change, but the study was only 14 weeks in duration.

Q: Glucokinase is also expressed in the hypothalamus. Was their any brain exposure?

A: The blood to brain ratio is very small. The brain concentration is vey small. I don’t have the exact concentration, but it can be defined as poor.

Q: Were there any changes to lactate or LDL?

A: There were no changes.


Symposium: Glucagon – Renaissance of an Old Hormone


Roger Unger, MD (University of Texas Southwestern, Dallas, TX)

Using data from preclinical studies conducted in his lab, Dr. Unger (who was unable to attend the meeting and had to call in remotely) provided a compelling argument for the need to target glucagon suppression in the management of type 1 diabetes. In particular, he demonstrated that leptin (a known glucagon suppressor) administration alongside insulin therapy in a mouse model of type 1 diabetes normalized glucose levels and eliminated glycemic variability. Additionally, glucagon receptor knockout mice exhibited normoglycemia, even after exposure to streptozotocin, and only developed elevated blood glucose upon introduction of glucagon receptors via an adenovirus containing glucagon receptor cDNA. Based on these findings, Dr. Unger concluded that type 1 diabetes, at least in mice, cannot exist without glucagon, suggesting that glucagon suppression in addition to insulin therapy should form a core part of type 1 diabetes treatment. He foreshadowed that GABA (an oral agent that suppresses glucagon) could become the first oral treatment for type 1 diabetes.

Questions and Answers

Q: How does glucose get taken up by cells and metabolism restored to normal without both insulin and glucagon?

A: We are currently studying that. In mice, we performed OGTTs and used 13-C labeled glucose and mass spectrometry to trace glucose in the body. Again, this is ongoing work. What we do know is that glucose ends up being stored in the liver as glycogen. However, it is made from a three-carbon chain precursor. Where those precursors come from we don’t know yet. Our hypothesis is that the glucose is getting into the liver without insulin. Once it is in there, it is getting chopped up into three-carbon fragments. From there, it gets transformed into glycogen. Perhaps in the absence of glucagon, you don’t need insulin. That is our hunch. Our work will tell us whether that is correct.

Q: Hormone independent transport of glucose plays a larger role in rodents than in humans. Do you have any studies planned that will examine this in humans?

A: I don't see how we can study this concept in humans besides suppressing glucagon action and stropping insulin treatment. I don’t think we’d ever get permission to do that study.

Q: In some humans, glucagon secretin is lost alongside insulin secretion. Those patients, however, develop diabetes. How does this fit in with your theory?

A: Studies have suggested that alpha cells also exist in the fundus of the stomach. In these individuals, if they are using insulin therapy, you will not detect glucagon release from these cells until insulin treatment is stopped. If you stop insulin treatment, you will see glucagon.



Jens Holst, MD, PhD (University of Copenhagen, Copenhagen, Denmark)

Dr. Holst provided a broad overview of the data supporting the use of glucagon pharmacomodulation in the treatment of type 2 diabetes, expressing concern over the side effects associated with glucagon receptor antagonists and optimism for glucagon secretion suppressants and glucagon receptor agonists.

  • With regard to glucagon receptor antagonism, Dr. Holst highlighted data for two small molecule glucagon receptor antagonists presented at ADA 2011, Merck’s MK- 0893 and Eli Lilly’s LY2409021. He noted that while providing robust glycemic control, one or both of the candidates were associated with a delay in hypoglycemia recovery and increases in LDL, hepatic transaminases, and weight, all of which he found concerning and problematic if they turned out to be class effects. (For our coverage of these results, please see pages 94 and 99 of our ADA 2011 Report at Notably, during Q&A, a representative from Eli Lilly remarked that data would be presented later at the meeting for LY2409021 demonstrating no associated delay in hypoglycemia recovery or increases in LDL or weight with treatment (981-P, 1002-P).
  • Turning to the suppression of glucagon secretion as a treatment strategy, Dr. Holst demonstrated that the glucose lowering effect of GLP-1 is equally attributed to insulin secretion stimulation and glucagon secretion suppression. Because of the positive safety profile, weight effects, and lipid effects of GLP-1 receptor agonists, he believed that glucagon secretion suppression could form an attractive alternative to glucagon receptor antagonism for the treatment of type 2 diabetes.
  • To close his presentation, Dr. Holst highlighted the promise held by glucagon receptor agonism for the treatment of type 2 diabetes. In particular, he pointed to a study in DIO mice in which treatment with a dual GLP-1/glucagon receptor agonist led to superior weight loss, lipid lowering activity, and similar blood glucose lowering activity as a GLP-1 receptor agonist (Pocai et al., Diabetes 2009). Citing other preclinical studies as well as data from a small infusion study in healthy volunteers, he suggested that the weight loss effects of glucagon agonism may be due to both decreased food intake and increased resting energy expenditure. We note that several companies are currently developed dual GLP-1/glucagon receptor agonists, including Transition/Eli Lilly (TT-401; phase 1) and Zealand Pharma/BI (ZP2929; preclinical).



Bethany Cummings, PhD (University of California – Davis, Davis, CA)

After discussing studies in which leptin therapy ameliorated hyperglycemia in type 1 diabetes in mouse models, Dr. Cummings explored the potential of leptin therapy in type 2 diabetes in UCD-T2DM mice, and in closing briefly touched on its potential clinical applications. In UCD-T2DM mice, exogenous leptin administration normalized fasting plasma glucose by: 1) decreasing circulating glucagon concentrations, likely leading to decreased hepatic glucose production; 2) improving lipid metabolism; and 3) improving insulin sensitivity, likely mediated by decreases of endoplasmic reticulum stress signaling. Dr. Cummings noted that while a clinical study showed no significant improvement in insulin sensitivity with leptin therapy for obese subjects recently diagnosed with type 2 diabetes, the therapy could perhaps be useful in combination therapy. She briefly flashed clinical data for pramlintide/metreleptin as an example of potential combination therapy; we note that while the efficacy of the combination is compelling, Amylin/Takeda discontinued development of the drug in 2011 (see our August 25, 2011 Closer Look at Looking ahead, we would really love to hear researchers explore the potential use of leptin for weight maintenance…


Basic Science

Banting Lecture


Bruce Spiegelman, PhD (Harvard Medical School, Boston, MA)

In this year’s Banting Lecture, Dr. Spiegelman delivered a fascinating presentation on the work conducted by his lab to elucidate the biology of brown fat regulation and to develop brown-fat based therapeutics for the treatment of metabolic disease. After describing the different functions of brown and white fat in the body, Dr. Spiegelman reviewed a series of in vitro, animal, and human studies that demonstrated that: 1) a distinct form of UCP-1 expressing thermogenic adipocytes from classical brown fat exists in rodents (which he termed beige fat); 2) increased beige fat in rodents improves glucose intolerance; and 3) beige fat exists in humans and appears to be the predominant thermogenic adipocyte in adults. Backtracking momentarily, he highlighted results from mice studies linking the expression of the transcriptional co-activator PCG1-alpha to beneficial effects associated with exercise (i.e., mitochondrial biogenesis, prevention of muscle atrophy). Interestingly, he showed that increased expression of PCG1-alpha in the muscle of transgenic mice led to the browning of white fat into beige fat. Subsequently, his lab discovered that a secreted moiety (which was named irisin) cleaved from FNDC5 (a protein whose expression was induced by PCG1-alpha) was responsible for this browning. To examine the in vivo effects of irisin, an adenovirus containing FNDC5 was injected intravenously into mice, which resulted in increased expression of UCP-1 in subcutaneous fat and an improvement in glucose homeostasis. Dr. Spiegelman next demonstrated that the major effect of irisin was the promotion of beige fat precursor cell maturation rather than the transdifferentiation of white fat into beige fat. Finally, he concluded by discussing the development an FC-fusion irisin therapeutic in his lab that was shown to have a half-life of nine days and the ability (with a single injection) to reduce fasting glucose and insulin levels in HDF mice. He stressed, however, that the drug was yet not optimized and not nearly ready for clinical development – although it could serve as a useful research tool. We note that Ember Therapeutics in-licensed irisin from the Dana-Farber Cancer Institute earlier this year. For more information on Ember, please see the December 22, 2011 Closer Look at

  • Dr. Spiegelman opened his presentation with a brief description of the differences between white adipose tissue and brown adipose tissue. White adipose tissue stores energy in a single lipid droplet, has relatively low mitochondrial content, expresses no uncoupling protein-1 (UCP-1), and is pro-inflammatory in the context of obesity. He noted that PPAR gamma has been identified as a key regulator of white adipose tissue development. In contrast, Dr. Spiegelman explained that brown adipose tissue plays an anti-obesity, anti-diabetes, and anti- hyperthermia role in most mammals, including humans. Through high mitochondrial content and the expression of the mitochondrial protein UCP-1, brown adipose tissue is capable of dissipating chemical energy content in the form of heat. Several key regulators of brown fat development have been discovered to date, including PPAR gamma, PGC1 alpha, PCG1 beta, and PRDM16.
  • Dr. Spiegelman highlighted that the presence of brown fat in adult humans was “rediscovered” through the work of PET imaging in the field of oncology. When PET imaging was used alongside radioactive glucose to detect metastases, a ring of symmetric hot spots of glucose uptake were frequently observed that did not appear to be metastatic tumors. These hot spots were particularly apparent following exposures to cooler temperatures. Open biopsies were performed on several individuals, and these hot spots were identified as areas of brown adipose tissue. Based on these findings, Dr. Spiegelman remarked that most individuals likely had some stores of brown adipose tissue in their bodies. He indicated that major questions facing the scientific community today include: 1) what role does brown adipose tissue play in overall energy balance, and 2) are there ways in which to increase the amount and/or function of brown adipose tissue in the body?
  • Brown adipose tissue is derived from a separate lineage than white adipose tissue. Dr. Spiegelman detailed results from his lab that demonstrated that suppression of the protein PRDM16 transformed primary cultures of brown adipose tissue into muscle tissue. Furthermore, driving the expression of PRDM16 in muscle tissue resulted in the formation of brown adipose tissue. Altogether, these results demonstrated not only that PRDM16 is a key regulator of brown adipose development, but that brown adipose is derived from the same lineage as muscle, not white adipose tissue.
  • There are two separate types of brown adipose tissue in the body (classical brown fat and beige fat), and beige fat helps improve the metabolic health of animals. Following the work above, studies in his lab noted that pockets of brown fat still emerged in white adipose tissue under conditions of extreme cold or extreme beta-adrenergic signaling. These results suggested that a separate type of thermogenic, UCP-1 expressing adipose type existed that shared the same lineage as white adipose tissue. In a separate experiment, the transgenic expression of PRDM16 in mice adipose tissue resulted in increased formation of beige fat in white fat tissue stores. Classical brown fat stores were not altered. In comparison to high-fat fed control mice, these transgenic mice exhibited significant improvements in glucose AUC following a glucose challenge.
  • PGC-1 alpha may help regulate the benefits of exercise in mammals. PGC-1 alpha is a transcriptional activator that is found in higher levels in brown fat and red oxidative muscle. Its expression is increased during exercise in rodents, mice, and humans. In muscle, PCG-1 alpha stimulates mitochondrial biogenesis, glucose uptake, neuromuscular junction formation, angiogenesis, muscle fiber type switching, and fatting acid oxidation – many of the benefits associated with exercise. Most strikingly, expression of PGC-1 alpha prevented muscle atrophy in mice unable to move following the severing of their sciatic nerve. Thus, by expressing PGC-1 alpha in cultured muscle tissue, Dr. Spiegelman noted that an experimental system could be created (at least to some extent) in a petri dish to study the molecular mechanisms underlying exercise.
  • The beneficial effects of PGC-1 alpha on adipose tissue are imparted by the action of the hormone Irisin. Intriguingly, the expression of PGC-1 alpha in the muscle of transgenic mice led to increased levels of beige fat in white adipose tissue stores. Through a series of experiments, Dr. Spiegelman’s lab identified a secreted soluble molecule (which they named Irisin) that is cleaved off the muscular transmembrane protein Fndc5 that appeared to be responsible for this “beigeing” of white fat. Of note, Irisin is nearly 100% identical between different mammalian species, indicating a high-degree conservation and restriction in molecular changes to the compound. This compares to 85% homology in insulin and 90% homology in glucagon between humans and mice. Irisin was found to circulate in both mice and humans, and the levels of circulating Irisin increases with exercise.
  • Elevated Irisin levels were associated with improvements in glucose homeostasis. An adenovirus vector expressing full length Fndc5 was injected into the tail vein of mice. After uptake and expression of the vector in the liver, circulating levels of Irisin increased two- to three- fold at 10 days following the injection. Subsequently, increased patches of UCP-1 expressing cells were detected in the subcutaneous fat of these mice, suggesting a “beigeing” effect. In high-fat fed obese mice transfected with this vector, significant improvements in fasting plasma glucose and glucose tolerance were observed, indicating that even modest elevations in Irisin can have positive effects on glucose homeostasis.
  • Dr. Spiegelman pondered what other therapeutic benefits Irisin may provide beyond its impact on adipose tissue. He noted that exercise also impacts the brain, the liver, and skeletal muscle. Focusing on the brain, he highlighted that exercise helps induce neurogenesis and has been shown to benefit individuals with Parkinson’s disease and Alzheimer’s disease. Thus, Dr. Spiegelman was particularly excited about the potential applications of Irisin or other similar compounds as a treatment for neurodegenerative diseases, especially given that many individuals with neurodegenerative diseases are unable to effectively exercise. Similarly, he believed that the opportunity to impart the benefits of exercise through a drug would also provide substantial benefit to individuals with morbid obesity and paraplegia.
  • Dr. Spiegelman detailed the work of his lab to develop an Irisin-based therapeutic for the treatment of metabolic disease. In particular, his lab fused the Fc fragment of immunoglobulin to the N-terminus of Irisin to enhance the hormone’s stability. When cultured murine adipose cells were exposed to this compound, a significant upregulation in UCP-1 was observed. The half-life of the compound in the blood of mice was notable nine days, and a single injection in high-fat fed mice led to a significantly increased expression of UCP-1 as well as significant improvements in fasting glucose and fasting insulin. Dr. Spiegelman stressed, however, that this was a proof of concept molecule and that it was not close to human use.
  • Dr. Spiegelman discussed other work in his lab aimed at isolating and characterizing beige adipose cells. Clonal cell lines were derived from the stromal vascular faction of murine subcutaneous adipose tissue, 23 of which underwent adipose differentiation. After analyzing the gene expression in these cell lines, the lines were clustered and were found to fall into two separate groups. One of these groups clustered more closely to classical brown adipose cells, which were cloned separately. Dr. Spiegelman’s group believed that these cell lines represented precursors to beige adipose cells, and the other group of cell lines (the ones that did not cluster as closely to classical brown adipose cells) represented precursors to white adipose cells. Interestingly, it was demonstrated that the beige precursor cells only began to express UCP- 1 and other markers characteristic of brown adipose tissue when stimulated with cAMP (a thermogenic stimulus). Prior to cAMP exposure, the beige cells largely resembled the white adipose cells. These results suggested that beige adipose cells or their precursors could hide in a white adipose tissue like state, but could rapidly be converted into thermogenic cells when exposed to the proper stimulus.
  • Returning to the PET scan findings discussed at the beginning of his lecture, Dr. Spiegelman reviewed work conducted by his lab and others to show that these identified pockets of thermogenic adipose tissue were actually beige adipose cells. Biopsy samples from patients were obtained the PET scan detected hot spot areas. Unambiguously, it was shown that these adipocytes expressed beige adipocyte specific markers (CD137, Tmem26, Tbx1) previously identified in mice beige adipocytes, but none of the classical brown adipocyte specific markers (Ebf3, Eva1, Fbxo31). These results suggested that the predominant thermogenic adipocyte type present in adult humans is beige fat, not brown fat. Additionally, the findings provided evidence that adult human beige fat closely resembles murine beige fat, making murine beige fat a useful model with which to explore the biology of human beige fat. Dr. Spiegelman expressed great optimism that information from such studies in the coming decades would lead to the development of brown/beige fat-based therapeutics for diabetes prevention and/or treatment. He noted that that Irisin was a nice first candidate, but unlikely the only candidate that will emerge.
  • Finally, Dr. Spiegelman demonstrated that Irisin acts to induce the activation of beige adipocyte precursors into thermogenic beige adipocytes rather than convert white adipocytes into beige adipocytes. Using CD137 as a marker of beige preadipocytes, mouse white preadipocytes (low CD137) and beige preadipocytes (high CD137) were sorted. When the Fc-fused Irisin molecule was applied to the white preadipocytes, no effect was observed. In comparison, Irisin induced a markedly increased expression of UCP-1 and other markers of thermogenic adipocytes (Cox8b, Prdm16) in the beige preadipocytes. Thus, Irisin does not appear to cause transdifferentiation of white adipocytes, but rather encourages beige adipocyte precursor cells to mature down a preordained pathway.


Current Issues: Perspectives on Mitochondria Dysfunction in Insulin Resistance


Bret Goodpaster, PhD (University of Pittsburgh, Pittsburgh, PA)

Dr. Goodpaster took an evidence-based, chronological approach to supporting his claim that mitochondrial deficiency promotes insulin resistance. He opened by noting three points: 1) it is important to think about mitochondria based on their role as mediators of oxidative stress; 2) there are multiple modes of insulin resistance, which can develop through different pathways; and 3) evidence must be translatable to humans. Dr. Goodpaster mentioned early data showing that people with type 2 diabetes exhibit impaired fatty acid oxidation.. He emphasized the uncertainty in whether this impairment was due to suppressed oxidative activity or a change in preferred substrates to be oxidized. Dr. Goodpaster then reviewed subsequent studies that found that mitochondrial oxidation activity declines in people with obesity and is especially reduced in patients with type 2 diabetes. Notably, the combination of weight loss and exercise has been found to improve both mitochondrial oxidation capacity and insulin sensitivity in people with type 2 diabetes. Further data show that in obese people, increased mitochondria oxidation is the strongest predictor for improved insulin sensitivity. During the final portion of his presentation, Dr. Goodpaster explained the possible role of oxidative stress as a potential mechanism for how mitochondrial oxidative activity influences insulin resistance. In rodents, a high fat diet increases oxidative stress, which was linked to insulin resistance. Furthermore, antioxidant activity was associated with improved glucose uptake and mitochondria oxidase activity. Dr. Goodpaster concluded his talk by noting that mitochondrial deficiency may represent one of the several pathways which drive insulin resistance.

Questions and Answers

Q: It is tempting to think that coupled respiration is most important. What is known about uncoupled respiration or the proton leak? Does greater mitochondrial mass increase leak?

A: I don’t have an answer; we are looking into that right now. I think there is a difference in mitochondrial coupling that could be very important

Q: Is there any causal relationship between abnormal mitochondria causing insulin resistance?

A: In some instances I would say yes, data supports this relationship.

Q: There are papers describing how metformin might impair mitochondrial function; can you comment on insulin signaling in the liver?

A: I can’t comment on insulin signaling in the liver.

Q: Exercise will increase whole-body metabolism. Can exercise enhance beta cell function?

A: The real question is whether exercise can have systemic effects. Even the brain mitochondria increase with exercise, in addition to muscle mitochondria.



John Holloszy, MD (Washington University School of Medicine, St. Louis, MO)

Taking the opposing view to Dr. Goodpaster, Dr. Holloszy argued that mitochondrial deficiency does not mediate insulin resistance. During his presentation, Dr. Goodpaster argued that the reduction in oxidation of fatty acids due to a deficiency of mitochondria is hypothesized to cause insulin resistance. In response, Dr. Holloszy referenced papers which show that a reduction in muscle mitochondria concentration neither precedes nor causes the development of insulin resistance, and that the muscles of type 2 diabetes patients do not oxidize less fat than those of people without diabetes. He concluded with the controversial statement, “people should stop wasting money on this” because the hypothesis is “nonsensical.” The audiences’ opinions on this topic started off as being roughly evenly split, and ended up slightly in Dr. Holloszy’s favor. The issue is still far from clear, however, and most audience members raised their hands when asked whether they were “totally confused.”

  • According to Dr. Holloszy, Dr. David Kelley’s report (Kelley et al., Diabetes, 2002) that the muscles of patients with type 2 diabetes contain fewer mitochondria than those of healthy individuals ignited interest in the relationship between mitochondrial deficiency and insulin resistance. It led to the hypothesis that mitochondrial deficiency reduces the oxidation of fatty acids in muscles, leading to increased insulin resistance. This theory was corroborated by subsequent studies which found that insulin resistant obese individuals and patients with type 2 diabetes have a roughly 30% reduction in mitochondria compared to people without insulin resistance.
  • Dr. Holloszy warned that correlation does not imply causality. For the hypothesis to hold true, the mitochondrial deficiency should result in decreased fatty acid oxidation in the muscle and should precede and lead to insulin resistance.. Such outcomes are difficult to measure in humans because insulin resistance develops years before a clinical diagnosis. However, Dr. Holloszy noted that rodent models are viable tools for evaluating the hypothesis, since rodents on high fat diets predictably develop obesity and insulin resistance in a laboratory setting.
  • Dr. Holloszy cited numerous studies that found that mitochondrial deficiency does not precede insulin resistance. According to the rodent models of Turner (Diabetes, 2006) and Hancock (PNAS, 2008), a high fat diet increases, rather than decreases, muscle mitochondria while causing insulin resistance. Furthermore, Garcia-Roves (PNAS, 2007) found that raising the free fatty acid concentration in rodent muscles through direct injection also increased mitochondrial concentration. Dr. Holloszy agreed with Dr. Gracia-Roves’ proposed mechanism for this effect – that free fatty acids lead to increased PPARδ binding to the carnitine palmitoyltransferase 1 promoter, which increases mitochondrial biogenesis.
  • Dr. Holloszy then argued that mitochondrial deficiency does not lead to insulin resistance. Addressing the hypothesis that mitochondrial deficiency causes insulin resistance through decreased oxidation of fatty acids, Dr. Holloszy explained that the 30% decrease in mitochondria concentration found in type 2 diabetes patients should not impair the ability of muscles to oxidize fatty acid in a significant manner –the capacity of muscles to oxidize fatty acids far exceeds the amount required to supply energy in the resting state. When mitochondrial activity deteriorates enough to actually reduce oxidation, increased glucose uptake and insulin action, instead of reduced insulin action, was observed in mice models (Colbert et al., Nature, 1996; Felber et al., Journal of Bioenergetics and Biomembranes, 1987). Dr. Holloszy also noted the studies of Ritov (AJP - Endo, 2010) and Han (PLoS One, 2011), which both showed that knocking out key parts of the mitochondrial electron transport chain in rodent muscle tissue did not lead to insulin resistance.
  • Finally, Dr. Holloszy pointed out that fatty acid oxidation was not reduced in diabetes patients. He emphasized the Nair study (Diabetes, 2008), which showed that oxidative phosphorylation in Asian Indians with type 2 diabetes is the same as in non-diabetic Indians, and higher than in healthy European Americans. Dr. Holloszy also mentioned the studies of Ara (Nature, 2011), Larson (Diabetologia, 2009), and Boon (Diabetologia, 2007), which allshowed that diabetes patients have similar or elevated rates of fatty acid oxidation compared to people without diabetes.

Questions and Answers

Q: You have shown a series of studies in rodents, but few studies in humans. In humans, it is my understanding that insulin resistance leads to decreased mitochondrial activity.

A: I think such decreases in mitochondrial activity are temporary. High fat intake does indeed lead to a decrease in mitochondrial reduction, but a subsequent increase follows shortly.

Q: Is the definition of mitochondria dysfunction only ATP oxidation?

A: The mitochondria have many functions, but the most important one in this case is ATP oxidation. Nevertheless, resting muscle cells have a very low rate of oxidative metabolism. They use ATP for mainly maintenance functions, so you need minimal mitochondrial production for that.

Oral Sessions: Diabetic Dyslipidemia


Kathleen Brown, PhD (American Diabetes Association, Research Triangle Park, NC)

Dr. Brown described the results of a randomized, crossover study (n=9) in high fat fed mongrel canines that looked at the effects of the selective GPR119 agonist GSK1292263 on lipid, cholesterol, and fatty acid endpoints. 6 mg/kg of GSK1292263 or vehicle was administered, and study endpoints included fasted glucose, total cholesterol, HDL, triglycerides, glycerol, free fatty acids, and body weight. Total cholesterol and HDL cholesterol were increased for the four-week treatment period (21.5% and 16.4%, p=0.0008 and 0.0025, respectively). Furthermore, reduced triglyceride levels were observed following a two meal tolerance test, although there was no effect on fasting plasma triglyceride levels. Overall, we are excited to hear additional encouraging GPR119 agonist data, and look forward to seeing how these data translate to humans. We note that as of February 2012, GSK1292263 was no longer listed in GlaxoSmithKline’s development pipeline. For more information on GPR119 agonists, including the competitive landscape, please see page six of our GTCbio 2012 Day #2 report at

Special Lectures and Addresses: National Scientific and Health Care Achievement Awards Presentation and Outstanding Scientific Achievement Award Lecture


David Altshuler, MD, PhD (Broad Institute, Boston, MA)

Dr. Altshuler presented, to a packed audience, his work using genome-wide association studies (GWAS) to identify 54 novel genetic variants commonly associated with type 2 diabetes and suggestions for how to best utilize data gleaned from GWAS for disease intervention. He asserted that attempting to use these genetic variants for prediction and prevention could be futile given the complex genetic basis for type 2 diabetes. Instead, he advocated for identifying loss-of-function mutations that confer protection against diseases to identify new drug targets, given that drugs more commonly inhibit rather than activate. Finally, he proposed that GWAS could be incorporated more routinely into the traditional drug development process to avoid expensive investigations into questions that can be answered by existing variation in human genetics.

  • In collaboration with Dr. Leif Groop’s (Lund University, Malmo, Sweden) team, Dr. Altshuler’s group discovered 54 new gene variants commonly associated with type 2 diabetes using genome-wide association studies (GWAS). The majority of these had not previously been identified; their identification has increased the fraction of variability in diabetes that can be explained genetically by about 10%. Dr. Altshuler also reported that GWAS has been used extensively in studying lipid disease; for example, new variants in HMG-CoA reductase, the rate-limiting enzyme in the cholesterol synthesis pathway, were discovered using GWAS.
  • Despite researchers’ great capacity to identify these genetic variants, following up with actionable indications is a challenge for three reasons: 1) most variants found by GWAS are found in regions of the DNA that do not encode proteins, so it is difficult to identify the function of each region and what specific gene(s) they affect; 2) since these data are purely observational, it is impossible to use these studies alone to investigate the effect of perturbing the function of these gene regions; and 3) there is a general lack of information about these gene regions because few had previously been implicated in type 2 diabetes.
  • Genetic variants discovered from GWAS can help us learn about the pathophysiology of diseases in order to better inform therapeutics. Conventionally, determining the genetic basis for disease has been thought to be useful in prediction and prevention. However, the genetic basis for diabetes is so complex that accurate prediction is difficult. Dr. Altshuler argued that we should focus instead on using GWAS to develop treatments, and since most drugs are inhibitory rather than activating, the most direct route in therapy development is identifying loss-of-function mutations that are protective against diabetes and have no other adverse effect. As an example, Dr. Altshuler highlighted the gene PCSK9, whose loss of function is protective against coronary artery disease; people who naturally have a homozygous loss of PCSK9 are healthy and fertile, suggesting that pharmacologically inhibiting it would be safe. In diabetes, a stop codon in the gene SLC30A8 has been identified to be potentially protective against diabetes, and studies are currently underway to confirm that its loss of function does not cause other adverse effects.
  • Dr. Altshuler proposed that GWAS information can also be used to re-evaluate current approaches to drug therapy development. As an example, he cited studies investigating whether raising HDL levels reduces the risk of heart attack. He pointed out that three such drug candidates in development that raised HDL failed to lower cardiovascular risk. Instead of spending “billions of dollars” on more drug trials, Dr. Altshuler suggested using GWAS to identify a gene modulating levels of HDL and examine the risk of heart attack in populations with naturally high levels of HDL. Such a study was done, and demonstrated that there was no difference for risk of heart attack for people carrying the high-HDL variant; this resulted in at least two companies stopping development of HDL drugs.


Symposium: Inflammation in Type 2 Diabetes and Results of the TINSAL-T2D Trial


Allison Goldfine, MD (Joslin Diabetes Center, Boston, MA)

Dr. Goldfine presented the results of Stage 2 of TINSAL-T2D, a trial evaluating the efficacy and safety of the non-steroidal anti-inflammatory drug (NSAID) salsalate as a treatment for type 2 diabetes. Overall, the results were not very exciting. Overall, the results were interesting and support the concept that an anti-inflammatory approach can have a benefit on A1c albeit not as great as other mechanisms. After 48 weeks, salsalate treatment (n=146) resulted in a modest but significant A1c reduction (0.24%) beyond placebo (n=140) from a baseline of 7.7%; Dr. Goldfine suggested that the magnitude of the difference was smaller than originally anticipated, but could likely be explained by the changes in concomitant medications in the two arms. Of note, at 24 weeks just prior to the changes in concomitant medications the A1c reduction was 0.5-0,6%. We also note the very low baseline A1c, which likely had an effect on the magnitude of the reduction in A1c. For the highest quartile patients (A1c 8.2-10.7) A1c was lowered by >0.6% at 48 weeks. Dr. Goldfine explained that the salsalate treatment conferred anti- inflammatory effects, as evidenced by lower white blood cells (p <0.001), neutrophils (p=0.003), and lymphocytes (p <0.001), and significantly increased adiponectin; there was no significant change in CRP. Salsalate increased fasting insulin and decreased C-peptide, which Dr. Goldfine believed is likely due to changes in insulin clearance. There was a trend toward increased systolic blood pressure with drug treatment, and significant increases in total cholesterol (6.6 mg/dl) and LDL (8.2 mg/dl) compared to placebo (p <0.001). Salsalate increased urinary albumin, which reversed following discontinuation. Mild hypoglycemia was more common with salsalate treatment, and salsalate use was associated with a modest increase in weight (~1 kg [~2 lbs]) (p<0.001) – both of these could be major negatives. All episodes of hypoglycemia were associated with concomitant SFU. As expected, tinnitus occurred more frequently with salsalate. In the TINSAL-FMD substudy, which assessed flow-mediated vasodilation as a marker of cardiovascular health, no changes in endothelial function were observed. Dr. Goldfine noted that the longer-term effects of salsalate need to be better assessed before using it as a treatment for type 2 diabetes. We doubt that there will be much pursuit of this.

  • In Stage 2 of TINSAL-T2D, patients were randomized to placebo (n=140) or 3.5 g/day salsalate (n=146). Individuals between the ages of 18 and 75 with type 2 diabetes and baseline A1c between 7.0% and 9.5% who were on diet and exercise or stable pharmacotherapy (up to three agents) were included in the study; patients who were on TZDs, insulin, and/or GLP- 1 receptor agonists were not eligible to participate in the study. Patients randomized to salsalate were started at a 3 g/day dose for the first two weeks, and escalated up to 3.5 g/day for the rest of the trial if tolerability wasn’t an issue. Changes in dosing on concomitant medications for diabetes, lipids, and blood pressure were avoided if possible over the first 24 weeks of the trial, then were adjusted based on good clinical practice, a questionable trial design decision in our view since we would assume this would make it more challenging to show the differences prompted by the main drug being studied. At baseline, participants had an average age of 56 years, diabetes duration of ~5 years, weight of 96.2 kg (211 lbs), BMI of 33.3 kg/m2, A1c of 7.7%, and fasting plasma glucose of 151 mg/dl. 5% were not on any medications, 41% were on monotherapy, 49% were on dual therapy, and 6% were on triple therapy.
  • After 48 weeks, salsalate treatment resulted in a modest but significant A1c reduction (0.24%) beyond placebo. Dr. Goldfine noted that the magnitude of the difference was a little smaller than originally anticipated, but could likely be explained by the changes in concomitant medications in the two arms. Intensity of concomitant type 2 diabetes medications decreased in patients on salsalate but increased in placebo. More patients taking salsalate achieved an A1c reduction ≥0.5% after 48 weeks – ~40% in the treatment arm versus ~22% in placebo. In addition, salsalate treatment reduced fasting plasma glucose by 11 mg/dl beyond placebo after 48 weeks of treatment (p<0.001).
  • The primary endpoint of the study was met in that salsalate lowered A1c following 48 weeks of treatment. The magnitude of the effect was less than hoped for but the results are confounded by the change in concomitant medications at 24 weeks and the low baseline A1c in the study. Despite these confounding issues salsalate, an anti-inflammatory, significantly reduced A1c. The real promise of an anti-inflammatory in type 2 diabetes is the potential to not only reduce A1c but also to have a positive effect on complications (both micro- and macro-vascular) associated with the disease. Other groups are working on anti-inflammatory approaches for type 2 diabetes and associated complications. Catabasis Pharmaceuticals is developing, CAT-1004, which is a conjugate of salicylate and the omega 3 fatty acid, DHA. CAT-1004 exhibits synergistic activity on the NF-kB pathway and in preclinical models of type 2 diabetes and inflammation and in the clinic. CAT-1004 recently completed Phase 1. In preclinical models of type 2 diabetes, CAT-1004 produced significantly greater efficacy on glucose lowering and glucose tolerance than salicylate alone. Catabasis believes this improved efficacy is due to mechanistic synergy produced on the NF-kB pathway and will translate to improved efficacy in patients with type 2 diabetes.

Questions and Answers

Q: Were there any differences in baseline characteristics that may have explained the modest A1c-lowering effect?

A: Patients in the two treatment arms were well matched for all of the parameters I showed. At the time of randomization, one-quarter of the patients randomized had baseline A1c below 7.1%. I think that could contribute to the diminished effect in magnitude in the whole group.

Q: The insulin effect is really intriguing. Is this in fact related to some of the cases of reducing concomitant medications? At what level do you think salsalate is modulating insulin clearance – the liver, or the periphery?

A: The change in insulin clearance was not seen in animal models, so it is hard to tease apart. It could be at the level of the liver; it could also be other places. It’s difficult to assess.

Q: What specifically drove the decreases in concomitant medications?

A: The increase in mild hypoglycemic events led physicians to decrease concomitant medications, whereas poor glycemic control led to increases in concomitant medications.


Steven Shoelson, MD, PhD (Harvard Medical School, Boston, MA)

Dr. Shoelson began with an overview of inflammation in obesity. He explained that lean adipose tissue differed from obese adipose tissue in both the number and type of macrophages present. In obese mice, adipose tissue had a greater number of macrophages and more M2 macrophages – the proinflammatory type. In lean mice, more M1 macrophages were present. However, he noted that macrophage type has been difficult to identify in human adipose. Immune cell concentration, including T and B cells, is also greater in obese mice. Next, Dr. Shoelson provided a brief introduction on salicylate and its anti-inflammatory properties. In particular, he discussed the results from stage 1 of the TINSAL- T2D study. After 14-weeks of treatment with salsalate (3-4 g/day), participants with type 2 diabetes achieved provide placebo-corrected reductions in A1c of 0.4-0.6% from a baseline of 7.7%. Additionally, treatment with salsalate led to improvements in fasting blood glucose, C-peptide, glucose utilization, free fatty acids, triglycerides, C-reactive peptide, and adiponectin. While Dr. Shoelson noted the exact mechanisms behind salsalate’s effects are unknown, NF-kB inhibition and decreased metabolic clearance rate of insulin are potential candidates. Also of interest, Dr. Shoelson pointed to mouse studies suggesting salicylate could moderate the inflammatory cascade leading to atherosclerosis.

  • Lean adipose tissue has different macrophage characteristics than obese adipose tissue. Dr. Shoelson showed pictures highlighting the difference between macrophage character in lean mouse vs. obese mouse. In lean mice, macrophages were scarcely present and individually located in adipose tissue. In obese mice, adipocytes typically had a ring of macrophages surrounding them. Dr. Shoelson hypothesized that this occurred because: 1) macrophages were responding to dead adipocytes; or 2) individual adipocytes were producing substances causing monocytes to hone in on them. Dr. Shoelson noted that either way, the cell would die once the macrophages targeted it. Moreover, Dr. Shoelson explained lean adipose tissue usually contained M2 macrophages, while obese adipose tissue typically contained M1 macrophages . Importantly, M1 macrophages are pro-inflammatory. However, he noted it has been difficult to identify macrophage type in humans and that there is a fair amount of controversy on this matter.
  • Leukocyte concentration increases in response to obesity. Compared to negative controls, mice on a high fat diet had greater concentrations of inflammatory immune cells in their adipose tissue including CD3 T cells, CD4 T cells, CD8 T Cells, and B cells. On the other hand, T regulatory cell concentration decreased. Dr. Shoelson noted that this was particularly interesting because T regulatory cells are anti-inflammatory.
  • Stage 1 of the TINSAL-T2D trial suggested lowering effects of salsalate on fasting blood glucose and A1c. Dr. Shoelson explained why they chose salsalate over other salicylic acid derivatives: salsalate is insoluble at stomach pH, is hydrolyzed and absorbed in the duodenum, does not change bleeding time, is generic and inexpensive, and has a long term safety profile in humans. At the three doses tested (3.0 g/d, 3.5 g/d, and 4.0 g/d), A1c, fasting blood glucose, and glycated albumin decreased over 14 weeks, whereas adiponectin increased (for more details, please see our ADA 2009 report).
  • While many mechanisms of action for salicylate have been proposed, Dr. Shoelson focused on NF-kB inhibition and the alteration of the metabolic clearance rate of insulin. In mice, obesity activated NF-kB in circulating monocytes, whereas salicylate inhibited NF-kB. Dr. Shoelson presented data from Goldfine et al. (CTS 2008) suggesting similar NF-kB responses to obesity and salsalate in humans. NF-kB activity in peripheral blood mononuclear cells decreased in obese humans with diabetes after salsalate treatment. In addition to NF-kB inhibition, Dr. Shoelson indicated that salsalate might also reduce the metabolic clearance of insulin. In individuals treated with high doses of salsalate, insulin levels remained high even when C-peptide levels decreased.
  • Salicylate could moderate the inflammatory cascade leading to atherosclerosis. In mice predisposed to developing atherosclerosis, salicylate reduced CD45 adhesion and infiltration in the aortic arches at two weeks. Dr. Shoelson was argued that if salicylate could be used to reduce inflammation, the cascade from obesity to atherosclerosis might be preventable.

Questions and Answers

Q: Not all obese patients are insulin resistant. Have you found any difference in M1 vs. M2 macrophages in the fat of these two distinct populations?

A: In our studies, we haven’t taken tissue samples to analyze fat. In TINSAL-T2D, we didn’t take fat samples, so we don’t have access to that. It would be nice to have samples, particularly of abdominal tissues, but those would be very hard to get from patients.



Marc Donath, MD (Universitätsspital Basel, Basel, Switzerland)

Dr. Donath presented on the role of IL-1 and IL-6 in people with type 2 diabetes. Beta cells have IL-1 receptors and produce IL-1β themselves. He suggested IL-1β plays a deleterious role in people with type 2 diabetes, as IL-1 antagonists have shown consistent success improving insulin secretion and glycemia. IL-6, on the other hand, appeared to have a beneficial effect in people with diabetes. Dr. Donath referred to a mouse model implicating IL-6 in mediating GLP-1 secretion in pancreatic alpha cells and intestinal L cells. He reminded the audience that IL-6 is one of the best predictors of type 2 diabetes. Notably, IL-6 also increases with exercise in mouse models. Dr. Donath proposed that IL-6 could serve different purposes. He suggested that in obesity, IL-6 is secreted from adipose to help the body respond to increasing insulin resistance and hyperglycemia. However in exercise, IL-6 may function to prepare the body for a post-exercise meal.

  • An elevated glucose concentration induces IL-1β. Beta cells are replete with IL-1 receptors and produce IL-1β in response to glucose. Dr. Donath suggested that the high expression of IL-1 receptors in pancreatic islets – more so than in all other tissues – indicates that these receptors play an inflammatory role in type 2 diabetes.
  • Eight studies have shown improved insulin secretion and glycemia with IL-1 antagonist intervention. Dr. Donath pointed to the overwhelming evidence suggesting IL-1 antagonists like canakinumab and anakinra (Amgen’s Kineret) confer improvements in patients with diabetes. He presented data from a gevokizumab study in people with type 2 diabetes in press for the July 2012 issue of Diabetes Care. Three months after a single injection of gevokizumab, participants showed a placebo-adjusted A1c reduction of 0.85% (the baseline was not available but this seems like a great response). Notably, gevokizumab also increased C- peptide secretion and decreased C-reactive protein levels.
  • Dr. Donath proposed that IL-6 reprograms pancreatic alpha cells to secrete GLP-1. Dr. Donath explained that proglucagon is processed into several end products, including GLP-1 and glucagon.. Pancreatic alpha cells produce prohormone convertase 1/3, which coverts proglucagon to GLP-1 in the intestine. In a rodent model, intermittently elevated IL-6 increased GLP-1 synthesis in both pancreatic alpha cells and intestinal L cells and increased. Prohormone convertase 1/3 levels in the pancreas. Dr. Donath suggested this pathway may be responsible for the glycemic benefits observed in IL-6-treated mice. In reviewing his previous work (Ellingsgaard et al., Nat Med, 2001), Dr. Donath noted that obese mice showed increased insulin levels and decreased blood glucose levels in an OGTT test when injected with a bolus of IL-6 compared to controls.
  • IL-6 levels are chronically elevated in obesity and intermittently elevated with exercise. Mice models indicate that exercise increases IL-6 and GLP-1 levels. Dr. Donath suggested that the increases in IL-6 during exercise and obesity result from distinct, non- conflicting pathways. Exercise-induced IL-6 secretion from skeletal muscle likely prepares the body for post-exercise meals while obesity-induced IL-6 secretion helps the body compensate for increased insulin resistance and hyperglycemia.

Symposium: Glucagon – Renaissance of an Old Hormone


Christine Longuet, PhD (University of Toronto, Toronto, Canada)

In an overflowing auditorium for the symposium on glucagon, Dr. Longuet argued that the glucagon receptor (Gcgr) is essential in lipid homeostasis and underscored that in rodent models, knocking out (KO) Gcgr leads to significant increases in hepatic triglyceride (TG) content and free fatty acid (FFA) production and secretion. Dr. Longuet found that KO mice had significantly elevated levels of plasma free fatty acids (FFA) (about 23.4 mg/dl; p <0.001) and TG (about 100 mg/dl; p <0.001) after a 16-hour long fast compared to WT mice (about 13.5 mg/dl and 30 mg/dl, respectively). Furthermore, KO mice were more susceptible to hepatosteastosis after exposure to a high fat diet (liver triglyceride content was about 11 mg/dl in KO mice but only about 5 mg/dl in WT mice; p <0.05) (Longuet et al., Cell Metabolism 2008). She commented that while there are very few studies looking at the impact of glucagon suppressors on lipid homeostasis, Xiao et al. (Diabetes 2011) found that glucagon inhibits hepatic lipoprotein secretion in humans. She indicated this finding suggests that her rodent studies may be translatable to humans.

Questions and Answers

Q: Do you think that the effect of glucagon on lipid metabolism would make it dangerous to use to treat diabetes?

A: Data in humans is still very preliminary. There is no guarantee that the phenotype in humans will be the same as it is in rodents. The impacts of glucagon suppression may be worse in the treatment of type 2 diabetes because it is already associated with lipodemia.

Q: How can we overcome hypoglycemia when suppressing glucagon?

A: We did not look at any way of overcoming that. I am not sure. I am sorry.

Q: Has anybody measured fatty acid metabolites?

A: Not that I know of.

Q: There is a case report of a human with an inactive glucagon receptor, and no liver abnormalities were reported. Wouldn’t this speak against your hypothesis that there should be this deleterious effect if you block Glcr?

A: These results were obtained in animal models, mainly rodents. We are not trying to say that this is what will happen in humans. But we need to monitor this in humans to make sure that this does not happen.

Symposium: Inflammation and Insulin Resistance


Steven Shoelson, MD, PhD (Harvard University, Cambridge, MA)

Dr. Shoelson opened the session on inflammation and insulin resistance with an overview of the epidemiological, biochemical, pharmacological, and immunological research to date that has helped characterize this relationship. Relevant drug targets he discussed were TNF-alpha, salicylates, and IL-1. Overall, Dr. Shoelson appeared ambivalent toward the applicability of TNF-alpha to metabolic disease therapy, presenting both positive and negative studies using TNF-alpha blockers. He expressed greater excitement for salicylates, noting that he was one of the lead investigators for the TINSAL and TINSAL-T2D trials. As a reminder, TINSAL-T2D was a clinical trial that examined the use of salsalate (a prodrug to salicylate) to treat type 2 diabetes. In the first stage of the study, salsalate (3-4 g/day) was demonstrated to provide placebo-corrected reductions in A1c of 0.4-0.6% from a baseline of 7.7% in people with type 2 diabetes over 14 weeks. Additionally, treatment with salsalate led to improvements in fasting blood glucose, C-peptide, glucose utilization, free fatty acids, triglycerides, C-reactive peptide, and adiponectin. Results from the second stage of the trial (a 52-week phase 3 study examining the efficacy and safety of salsalate in people with type 2 diabetes) were presented by Dr. Shoelson at a Monday symposium at this year’ s ADA entitled “Inflammation in Type 2 Diabetes and Results of the TINSAL-T2D Trial.” Dr. Shoelson commented that salsalate could serve as a valuable therapy for type 2 diabetes given its low cost and ability to target several of the inflammatory components of diabetes. Finally, Dr. Shoelson closed his talk with a brief overview of the role of IL-1 in type 2 diabetes pathogenesis. In particular, he highlighted positive results from a study by Larsen et al. (NEJM 2007) that found IL-1 antagonism with anakinra (Kineret) in people with type 2 diabetes resulted in reductions in A1c (-0.46% placebo-adjusted), improvements in beta cell secretory function, and reductions in systemic markers of inflammation (IL-6 and C-reactive protein). As a reminder, XOMA’s anti-IL-1 beta antibody XOMA 052 exhibited weak glycemic control efficacy (marginal, non-significant reductions over placebo), but provided significant reductions in C-reactive protein in both a six-month phase 2a and six-month phase 2b trial in people with type 2 diabetes in 2011 (see the June 10, 2011 Closer Look at


Jenny Ting, PhD (University of North Carolina, Chapel Hill, NC)

Dr. Ting discussed innate immune receptors and the inflammasome and their relevance to metabolic disease. Her research has focused on nucleotide binding-leucine rich repeat proteins (NLRP), which respond to a variety of signals by activating inflammasome genes. There are at least seven inflammasome genes, the protein products of which form multi-protein complexes (called inflammasomes) that contribute to cell death through the activation of caspases (particularly caspase 1) and the secretion of pro-IL-1B and IL-18. Dr. Ting suggested that the secretion of IL-1 from inflammasomes could induce insulin resistance in surrounding tissues by inhibiting the action of IRS1 on Akt. Fatty acids appear to be directly involved in this pathway by inhibiting AMPK in macrophages, which in turn activate the inflammasome pathway. The pharmacologic agent AICAR reverses the deactivation of AMPK in response to fatty acids, and in in vitro studies, adding AICAR to cell cultures reduced IL-1 secretion and inflammasome activation in response to fatty acids. Dr. Ting believes that a deeper understanding of the inflammasome pathway could help elucidate the mechanisms and effects of a number of metabolic pharmaceuticals. Delineating how IL-1 affects insulin sensitivity, she cultured macrophages with fatty acids and an inflammatory promoter (lipopolysaccharide), then exposed this medium to hepatocytes. In response to this inflammatory environment, the hepatic cells responded with a loss of nearly all insulin sensitivity as measured by Akt phosphorylation. Knocking out NLRP genes ameliorated this effect. Providing further evidence for the role of inflammation in the development of insulin resistance, she demonstrated that the knockout of IL-1 and TNF-alpha improved glucose control in mice fed a high-fat diet.



Mark Febbraio, PhD (Baker International Diabetes Institute, Melbourne, Australia)

Dr. Febbraio discussed his research involving macrophages, obesity, and insulin resistance. In severe metabolic syndrome and obesity, macrophages are typically observed engulfing apoptosing adipocytes in the histology of adipose tissue. Dr. Febbraio suggested that this occurs due to a positive feedback loop between overgrown adipocytes that secrete inflammatory cytokines and macrophages, eosinophils, and other immune mediating cells that are recruited toward these adipocytes, which in turn release additional inflammatory factors. To further investigate macrophages’ role in insulin resistance, Dr. Febbraio engineered a mouse knockout of CD36, a lipid transporter found primarily on macrophages. He and his colleagues observed a decreased inflammatory response to a high-fat diet with this genetic alteration. Macrophages did not accumulate in adipose tissue, and there was superior insulin signal transduction in adipocytes. To block macrophage chemotaxis therapeutically, he highlighted IL-6 antagonism as an approach that may be effective. IL-6 is a mixed pro- and anti-inflammatory cytokine involved in systemic and local inflammation. It plays a particularly important role in recruiting immune cells to areas of inflammation. Tocilizumab (Roche) is an antibody that neutralizes IL-6 and is currently used to treat rheumatoid arthritis. However, Dr. Febbraio noted that IL-6 antagonism has been associated with increased triglyceride levels, weight, and insulin resistance. Dr. Febbraio suggested that scavenging and deactivating a particularly active compounded form of IL-6 (IL-6 bound to a soluble receptor, IL-6R), rather than antagonizing the lone cytokine itself, could reduce macrophage recruitment to areas of inflammation while avoiding these metabolic side effects. To explore this hypothesis, Dr. Febbraio developed a mouse model that was genetically engineered to express a soluble receptor for the soluble IL-6/IL-6-receptor complex, which binds to and removes the complex from circulation. The introduction of this protein resulted in normal adipose tissue histology in high-fat-diet mice as well as greater peripheral insulin sensitivity in skeletal muscle. Interestingly, there was no effect on the insulin sensitivity in adipose tissue as well as the liver. Dr. Febbraio concluded by commenting that the next evolution in IL-6 therapy should involve developing ways to block the IL- 6/IL-6-receptor complex rather than just the IL-6 receptor itself.


Dongsheng Cai, MD, PhD (Albert Einstein College of Medicine, Bronx, NY)

Dr. Cai discussed the role of NF-kB expression in the hypothalamus in the pathology of type 2 diabetes, obesity, and hypertension. He explained that the hypothalamus was a logical area in which to investigate this connection, since it releases hormones that affect a wide range of metabolic and cardiovascular functions. NF-kB activates an inflammatory cascade within a cell that leads to changes in cell morphology, secretion of and response to immune mediators, and a wide range physiological functions. Typically, NF-kB is released in response to stressors such as metabolite excess, oxidative stresses, ER stress, and autophagic defects. In a study conducted in his lab, active IKK-beta/NF-kB was injected into the hypothalamus of rats. The rats subsequently developed hypothalamic neuronal insulin resistance. Dr. Cai indicated that under normal conditions, hypothalamic neurons respond to increases in insulin by reducing hunger and increasing peripheral metabolism. Conversely, suppressing the expression of hypothalamic IKK-beta/NF-kB in high-fat fed rats led to reduced weight gain, an anti- aging effect, an improvement in memory, and increased longevity (average lifespan of 1200 days vs. 1000 days in the control group). Dr. Cai hypothesized that a decrease in sympathetic output (which raises heart rate, blood pressure, and metabolism) triggered by the decreased levels of IKK-beta/NF-kB and inflammation in the hypothalamus might have contributed to these findings. He revealed that a measurable decrease in blood pressure was detected in the study, providing some evidence to support his hypothesis.

Questions and Answers

Q: What do you think is driving this increased inflammation in the hypothalamus? Have you repeated your studies in specific cytokine-knockout mice?

A: We haven’t studied these specifically, but the cytokine knockouts could help to elaborate the key players.

Q: In the human trials examining salicylate, there was an improvement in glucose and triglyceride levels, but there was no effect on hypertension. How do you rectify that with your data?

A: That is indeed a discrepancy; it may be related to the animal model as opposed to humans.

Q: Have you tried to study the downstream pathways in metabolic syndrome?

A: Downstream is very broad, particularly when we’re talking about the brain. We have some other studies with nutrient sensing and exocytosis of peptide neurotransmitters. But in terms of immediate downstream, we have not done much in that regard. We try to focus on NF-kB itself.

Q: Did you look at APK activity in your animal models in the hypothalamus?

A: No, we didn’t.

Symposium: Epigenetic Control of Insulin Action


Allan Vaag, MD, PhD (Copenhagen University, Copenhagen, Denmark)

Dr. Vaag presented evidence that insulin resistance may be under substantial epigenetic control. Epigenetics are the changes in gene and protein expression beyond what is encoded in the primary DNA sequence. For example, methylation of gene promoters can silence their transcription and miRNA expression can prevent the translation of protein from mRNA. Insulin resistance in type 2 diabetes is associated with impaired mitochondrial function in muscle, and data show that methylation of PGC-1α, a gene necessary for the synthesis of mitochondria, is increased after bed rest and after five days of overfeeding on a high fat diet. This increase in methylation was reversed after returning to a normal diet. Furthermore, the association between low birth weight (LBW) and risk for type 2 diabetes may be explained by differences in PGC-1α methylation: LBW individuals have a higher rate of PGC-1α methylation and do not respond to changes in diet (high fat or normal) by increasing or decreasing the methylation status of PGC-1α. Dr. Vaag concluded that the ability to respond to overfeeding with changes in DNA methylation may be protective against insulin resistance in normal individuals. Furthermore, LBW is associated with increased expression of various miRNAs that play a role in fat deposition and the insulin signaling pathway. Thus, LBW may be associated with type 2 diabetes via the epigenetic changes that occur in utero when the fetus is provided with insufficient nutrition.

  • Insulin resistance in type 2 diabetes is associated with impaired mitochondrial function in muscle tissue. This finding suggest that the differential regulation of genes involved in muscle mitochondrial function may explain the development of insulin resistance. Genes expressing mitochondrial proteins involved in oxidative phosphorylation are found to be expressed at a lower level in people with type 2 diabetes and in insulin-resistant first-degree relatives.
  • Increased methylation of PGC-1α, a gene necessary for the synthesis of mitochondria, in response to physical inactivity and high fat diet may help explainwhy these two habits lead to muscle insulin resistance. Methylation of genes at CpG sites in their promoter regions is typically associated with decreased gene expression. A mutation introducing a CpG site into the PGC-1α promoter region would result in increased methylation and decreased expression of the gene. It has been shown that nine days of bed rest leads to a significant increase in insulin resistance, and Dr. Vaag presented data demonstrating that bed rest was associated with increased methylation of PGC-1α close to the transcription start site of the gene. Furthermore, a study which fed a high fat diet to 11 young men for five days led to significant increases in methylation at two of the same four CpG sites, which was reversed by returning to a controlled diet. An examination of several other genes found that multiple CpG sites (25% of the ones studied) in human muscle biopsies exhibited altered methylation patterns (p<0.05) during the five days of high fat overfeeding, but these were very modest changes (most increased by only 2%, and none by more than 10%).
  • DNA methylation could explain the association between low birth weight (LBW) and risk for type 2 diabetes due to LBW individuals’ reduced ability to adjust their DNA methylation patterns when exposed to short-term, high-fat feeding. A 1 kg increase in birth weight translates to a ~45% reduction of diabetes risk. PGC-1α was methylated to a greater degree in LBW participants, but a high fat diet did not result in a subsequent increase in PGC-1α methylation as it did for their normal birth weight counterparts. Furthermore, an array study examining methylation of multiple genes after exposure to high-fat overfeeding showed that LBW individuals responded with only a 5.7% difference in DNA methylation, whereas normal birth weight individuals responded with a 25.3% difference in DNA methylation (p<0.001). Based on these findings, Dr. Vaag concluded that the ability to show flexibility in gene methylation in response to overfeeding may be protective against insulin resistance in normal individuals.
  • LBW is associated with increased expression of miR-483 in fat (which influences adipose tissue expandability and lipotoxicity) and with increased expression of miR-15b and miR-16 in muscle, which affects insulin signaling. miRNAs target complementary mRNA sequences, preventing the translation of the encoded protein. miR-483’starget mRNA plays a role in the proliferation of subcutaneous tissue, thus an increase in miR-483could result in the inability to store fat subcutaneously. This increases the storage of fat in the liver or muscle tissues, contributing to insulin resistance. Additionally, LBW has been associated with increased expression of miR-15b and miR-16. The most prominently predicted target for the miR-15 family is the insulin receptor pathway.

Questions and Answers

Q: Have you done any experiments to demonstrate that methylation actually reduces the activity of PGC-1α? For example, you could introduce methylation sites into the promoter and measure luciferase activity – then you could actually conclude that increased methylation changes PGC-1α activity. PGC-1α is very complicated, and many things regulate its expression.

A: That is a wonderful question. We do not have this data.

Q: My impression from a basic molecular biology standpoint is that DNA methylation is a relatively stable, long term change, so seeing such short term changes is astonishing. Do you know what is regulating these relatively rapid changes?

A: This is what I am most fascinated about. I think we should be very cautious not to conclude too much because the changes we see are small and there are some difficulties in working with these tissues. There are a lot of issues to address here.

Q: You stated that the change in methylation status was quite small in response to overfeeding – is there any hope of seeing this translate into biologically significant changes in expression?

A: We overfed the participants for only five days - what if you overfed people for a year, 10 years, or 20 years? At that stage, gene expression might change. Perhaps the changes right now are too small to see. Additionally, this was all done in resting muscle, and I think it is a bit naïve to think we can notice changes in this state. I think the muscle needs to be active – we need to challenge the muscle.

Symposium: Insulin Action via the Brain


Christoph Buettner, MD, PhD (The Mount Sinai Hospital, New York, NY)

Dr. Buettner discussed his work examining the role of central insulin signaling in the regulation of adipose tissue lipolysis and lipogenesis in a rodent model. An important distinction was first made between hepatic glucose production (HGP) and lipolysis: HGP is a somatostatin-dependent phenomenon, whereby blood glucose levels are only affected when insulin and somatostatin are dosed. In contrast, lipolysis is controlled by brain insulin independent of somatostatin. Two inducible insulin receptor knockout mice were used to test the effects of insulin on lipolysis: 1) a peripheral insulin receptor knockout, and 2) a whole body knockout. In the peripheral knockout, brain insulin was still able to suppress lipolysis, while in the whole body knockout, insulin could no longer suppress lipolysis. Furthermore, the whole body knockout was accompanied by a marked increase in several pro- inflammatory markers, while the peripheral knockout exhibited anti-inflammatory effects. Overall, Dr. Buettner commented that it appeared clear that brain insulin has direct effects on fat (at least in animals); however, the physiological importance of this relationship in humans remains to be explored.

Symposium: Novel Concepts in the Pathophysiology of Hypoglycemia in Diabetes


Stephen Davis, MBBS (University of Maryland School of Medicine, Baltimore, Maryland)

Dr. Davis discussed mechanisms for hypoglycemia-associated autonomic failure, mechanisms for exercise-associated autonomic failure, and approaches to enhance autonomic nervous system and neuroendocrine responses during hypoglycemia. Though much is still not known about the potential mechanisms involved, there are several novel approaches to amplifying the counterregulatory response during hypoglycemia. Dr. Davis noted that while these approaches are still under investigation, they demonstrate early promise.

  • Individuals who take GABA agonists (GABA being an important neurotransmitter in the central nervous system) before experiencing hypoglycemia the next day experience autonomic failure and a reduction in endocrine responses. Taking aGABA-a agonist and experiencing hypoglycemia in the same day leads to an increased glucagon response the following day, indicating the complexity of the pathways of endocrine responses and how much is still not understood.
  • People experience a similar reduction in endogenous glucose production, epinephrine, glucagon, and direct sympathetic activity when they have an episode of hypoglycemia after exercising. Cortisol infusions antecedent to exercise also significantly reduced epinephrine, glucagon, norepinephrine, and pancreatic polypeptides.
  • A number of approaches to enhancing these responses have been identified. Though one week of troglitazone therapy resulted in hepatic toxicity, there is a significant increase in both epinephrine and glucagon during subsequent hypoglycemia. Catalytic amounts of fructose have been found to increase epinephrine and endogenous glucose production. Administration of Dehydroepiandrosterone (DHEA) preserves the glucagon and epinephrine response during repeated hypoglycemia . Dr. Davis discussed plans to perform clinical trials to investigate the effects of DHEA on hypoglycemia.


Owen Chan, PhD (Yale University, New Haven, CT)

Dr. Chan’s presentation examined both the intra-islet and central nervous system (CNS) factors affecting glucagon secretion as they relate to diabetes. He also emphasized the importance of the neurotransmitter gamma-aminobutric acid (GABA) in associating CNS and peripheral glucagon regulation. While precise mechanisms for the relationship between insulin and glucagon regulation at both the CNS and periphery have yet to be entirely elucidated, Dr. Chan provided an excellent analysis of this complex network and the current understanding of these regulatory pathways.

  • Dr. Chan reviewed the intra-islet hypothesis, which purports that insulin release during increasing glucose levels can inhibit glucagon secretion from alpha cells. Additionally, Dr. Chan described how the ventromedial hypothalamic nucleus (VMH) is the primary glucose sensor in the brain – glucose delivery into the VMH suppresses the counterregulatory responses during hypoglycemia. In diabetes, glucagon release is determined by the net effect of both CNS and peripheral inputs to the alpha cell.
  • Dr. Chan also emphasized the importance of the neurotransmitter gamma- aminobutric acid (GABA) in associating CNS and peripheral glucagon regulation. GABA receptors are expressed on alpha cells, with GABA acting to inhibit glucagon secretion. Interestingly, insulin has been shown to increase GABA receptor translocation, illustrating another pathway through which insulin and glucagon regulate each other.

Symposium: Protein Acetylation and Other Post Translational Modifications as Metabolic Sensors


Matthew Hirschey, PhD (Duke University Medical Center, Durham, NC)

In this fascinating talk, Dr. Hirschey discussed a single nucleotide polymorphism (SNP) in the SIRT3 gene that has been implicated in metabolic syndrome. Participants in the Non-alcoholic Stetatohepatitis Clinical Research Network (n=834) and the Metabolic Syndrome in Men study (n~8000) were found to be at a statistically significantly higher risk of developing metabolic syndrome if they possessed this particular SNP in SIRT3. Dr. Hirschey therefore suspects that this SNP inhibits SIRT3, which leads to metabolic syndrome via decreased metabolic rate and fatty acid oxidation.

  • SIRT3 belongs to a class of proteins called sirtuins that regulate acetylation of mitochondrial proteins, especially those involved in fatty acid metabolism and the generation of energy. Chronic high-fat diet feeding suppresses SIRT3, but the over-expression of PGC-1 alpha, an important metabolic cofactor, rescues SIRT3 expression during high-fat feeding. Dr. Hirschey has found that mice that lack the SIRT3 gene develop several metabolic complications, including obesity, diabetes, insulin resistance, hyperinsulinemia, inflammation, and fatty liver disease.
  • A single nucleotide polymorphism (SNP) in SIRT3 is correlated with increased risk for the metabolic syndrome. 12 SNPs in SIRT3 were interrogated in the NASH-CRN (Non- alcoholic Stetatohepatitis Clinical Research Network), which included 834 participants (mean age 48.9, 36:64 male:female, 34.6 kg/m2), of whom 30.6% had diabetes and 47.7% had the metabolic syndrome. rs11246020, a SNP in the SIRT3 gene, was found to be correlated with increased risk for the metabolic syndrome with p=0.008. The same SNP was implicated in higher risk of metabolic syndrome in the METSIM (Metabolic Syndrome In Men) study, which included approximately 8000 Finnish male participants. This SNP encodes a mutation in SIRT3 that significantly reduces enzymatic efficiency, inhibiting the deacetylating activity of SIRT3. Dr. Hirschey suspects that these effects result in decreased metabolic rate and fatty acid oxidation, leading to higher lipid content, insulin resistance, and obesity, culminating in the metabolic syndrome.
  • Dr. Hirschey then investigated the effects of high-fat diet feeding on other mitochondrial protein acylations, but more research needs to be done. High-fat diet feeding resulted in increased hepatic propionylation, which is similar to acetylation but adds an additional carbon atom. The relevance of propionylation to metabolism and metabolic disease remains to be seen.

Questions and Answers

Q: Are you, or is anyone in the field, trying to work on an algorithm to figure out which lysine residues are most likely to be acetylated?

A: The crystal structures of the sirtuins appear to be binding to the backbone rather than the specific amino acid side chain, so there might be promiscuity to the sirtuins, which is what a lot of people have seen. I know that here is some work going on right now to identify which sites are SIRT-3 dependent and which are not.

Q: As a clinician, we are thinking about using these signatures to phenotype patients and understand what’s happening to them. We see there is a negative energy balance and non- alcoholic fatty liver disease. But, if you put them all on unsaturated fat diets, liver fat goes down. What’s driving these enzymes? Is it the energy balance? Could this be used to assess the status of these proteins?

A: I’d say there is a paucity of work regarding the actions of SIRT3 on the transcriptional level. But it is not known if or how the lipids are regulating this. Thinking past SIRT3 and the regulation status of mitochondrial proteins – we’ve tried to do this in some human patient samples, but it’s difficult because to get good mitochondria, you’d have to purify immediately. The strategy on of my colleagues has used to make a site-specific acetylation utilizes antibodies which can read out if you had a panel of these antibodies. The panel could read out the acetylation status of a key number of proteins that would perhaps be indicative of the energy balance, but the specific sites that you would want to target antibodies against is still unclear.

Symposium: The Basic Science of Exercise−Implications for Diabetes


Christoph Handschin, PhD (University of Basal, Basel, Switzerland)

Dr. Handschin presented an in-depth analysis on the physiology of PGC-1α, focusing on its relevance to glucose response and the potential therapeutic benefits from modifying its expression. In conducting several studies on transgenic PGC-1α mice, his group showed that PGC-1α plays a significant role in muscular adaptations to exercise – specifically, it enhanced mitochondrial proliferation in response to endurance exercise. The same studies found that PGC-1α promotes intramuscular lipogenesis or lipolysis (depending on if the mouse was sedentary or active, respectively), partially by modulating the pentose phosphate pathway. PGC-1α also increased muscular glucose uptake, hypothetically to replace the metabolized intracellular fatty acids.

  • Dr. Handschin cautioned the audience at the end of his talk that PGC-1α may not be a beneficial drug to replace exercise in a sedentary population. Studies show that sedentary mice on high fat diets with elevated PGC-1α had poor glycemic profiles compared to mice with no PGC-1α increases. He attributed this to a PGC-1α-stimulated increase in fatty acid concentrations in the muscle cells. Dr. Handschin explained that exercise mimetics such as PGC- 1α may be effective when combined with exercise but may increase pathologies in sedentary populations eating Western diets. Still, some questions were raised regarding this conclusion in relationship to previously conducted studies, indicating a need for additional research.


Laurie Goodyear, PhD (Joslin Diabetes Center, Boston, MA)

Dr. Goodyear’s presentation summarized several studies that her team conducted by transplanting subcutaneous adipose tissue (AT) from exercise-trained and untrained mice into the visceral cavity of other mice. They found that transplanting AT from mice that had been exercise-trained greatly improved glucose tolerance compared to transplanting AT from untrained mice. Dr. Goodyear’s team also conducted a similar study in mice that were fed high fat diets before receiving the transplant. In these studies, mice that received the exercise AT transplant were able to reverse the impaired glucose tolerance induced by the high fat diet.

  • After transplanting subcutaneous adipose tissue (AT) from exercise-trained and untrained mice into untrained mice, Dr. Goodyear and her team concluded that there are significant changes that occur in AT along with exercise, and that these changes affect systemic glucose tolerance. The results from this transplantation show very apparent (>30%) glucose improvements at nine-days post transplant, but are negligible at 12 weeks. The mice that receive AT from exercise-trained mice had better glucose tolerance, lower insulin release, and increased brown adipose tissue activity. Dr. Goodyear noted that this study demonstrated that exercise has marked effects on AT that in turn influence systemic physiology.


Product Theater (Sponsored by Boehringer Ingelheim and Eli Lilly)


Mohammad Abdul-Ghani, MD, PhD (University of Texas Health Science Center, San Antonio, TX) and Carol Wysham, MD (University of Washington, Seattle, WA)

Drs. Carol Wysham and Muhammad Abdul-Ghani attracted a standing-room only crowd, despite the noise level of the exhibit hall, with a basic science presentation focusing on the mechanistic roles of SGLT-2 receptors, particularly in type 2 diabetes. Dr. Wysham began the presentation with an overview of the impact of glucotoxicity on beta cells and noted that people with diabetes carry a risk of cardiovascular events similar to people with cardiovascular disease and no diabetes. Dr. Abdul-Ghani then explained the role of SGLT-2 transporter, noting that glycosuria results when its capacity (about 180 mg/dl) is exceeded at normal transporter expression levels. However, SGLT-2 protein levels are increased in type 2 diabetes, as shown by the over-expression of renal SGLT-2 in rats with diabetes. Furthermore, studies of human exfoliated proximal tubule epithelial cells show increased SGLT-2 and GLUT2 transporter expression. Dr. Abdul-Ghani noted that in mouse models with a SGLT-2 gene deletion, the absence of SGLT-2 transporters dramatically lowered fasting plasma glucose by roughly 60-70 mg/dl (p <0.001). Furthermore, diabetic mice with this deletion showed improved beta cell function compared to control diabetic mice, as shown by a two-fold increase in insulin secretion under hyperglycemic clamp. Dr. Abdul-Ghani further explained that the benign spontaneous loss of SGLT-2 function in a rare condition called familial renal glycosuria provides evidence that SGLT-2 inhibition therapy will be safe.

-- by Hannah Deming, Jessica Dong, Benjamin Kozak, Kira Maker, Mark Sorrentino, Josh Tennefoss Tanayott Thaweethai, Nick Wilkie, Vincent Wu, Katrina Verbrugge, and Kelly Close