David Cummings, MD

Background:

Dr. Cummings received his AB in Biochemistry from Dartmouth College, and his MD from Harvard Medical School and the Massachusetts Institute of Technology.  He performed his residency and chief residency in Medicine at UW.  As a fellow at UW, he was mentored by Stan McKnight in Pharmacology and William Bremner in Endocrinology.  In addition to teaching at UW and seeing patients at the VA Hospital, he has maintained continuous NIH funding for 15 years, studying body weight regulation, obesity, and diabetes, and he has >100 publications in this field.  Dr. Cummings is Deputy Director of the UW Diabetes Endocrinology Research Center and is a member of the American Society for Clinical Investigation.  He received the U.S. Presidential Early Career Award for Scientists and Engineers, the highest award conferred by the US government to researchers in their early independent careers.  Given at the White House by President George W. Bush, it was accompanied by 5 years of funding.

Focus:

Dr. Cummings' research focuses on regulation of food intake, body weight, and glucose homeostasis, with particular focus on the gut-brain axis.

Mechanisms Mediating Resolution of Diabetes Following Gastrointestinal Surgery

Our work implicates impaired ghrelin secretion in the effects of Roux-en-Y gastric bypass (RYGB) on food intake, body weight, and glucose homeostasis.  We now seek to examine additional mechanisms by which RYGB rapidly and dramatically resolves type 2 diabetes, through actions beyond just reduced food intake and body weight.  We study animals with a gastric-sparing bypass of the amount of proximal small intestine that is bypassed in a typical RYGB.  This duodenal-jejunal bypass operation (DJB) markedly improves or resolves type 2 diabetes in rodents and humans, independent of changes in food intake or body weight, and without causing malabsorption.  Our collaborative work with Francesco Rubino implicates the exclusion of nutrients from the duodenum as an important component of the anti-diabetic effects of DJB, rather than just increased GLP-1 secretion from expedited delivery of ingested nutrients to the distal intestine.  Glucose tolerance also improves in animals and humans following implantation of a plastic endoluminal sleeve into the proximal small intestine, preventing contact between ingested food and the duodenal lumen. 

Our current and proposed studies seek to elucidate the anti-diabetic mechanisms of duodenal bypass with experiments in rats, pigs, and humans.  We have worked with David Flum to develop and characterize porcine models of gastric bypass and DJB, which we use for studies requiring large animals.  Using insulin-resistant Ossabaw pigs, we are characterizing the effects of various surgical manipulations of the gastrointestinal tract on the details glucose homeostasis.  We will perform complementary studies in rats, collaborating with Mike Schwartz, Greg Morton, and the Animal Physiology Core, to determine whether proximal intestinal bypass (achieved with a DJB or duodenal sleeve) ameliorates diabetes by increasing insulin secretion, insulin sensitivity, or both.  These experiments involve frequently sampled IV glucose tolerance tests and hyperinsulinemic-euglycemic clamps in various rat models of diabetes.  In February 2009, we received a fundable score on a new RO1 grant proposing complementary human studies.

Roles and Mechanisms of Action of Ghrelin in the Regulation of Food Intake, Body Weight, and Glucose Homeostasis

For several years we have studied the actions and regulation of ghrelin, a relatively recently discovered orexigenic (appetite-stimulating) peptide hormone that is produced primarily by the stomach and proximal small intestine, with lesser expression in pancreatic islets.  Our areas of study are as follows.

Regulation of Circulating Ghrelin Levels 

The following is summarizes some of the findings from our ongoing studies to clarify mechanisms of ghrelin regulation.  We demonstrated that plasma ghrelin levels surge shortly before each meal, implicating this orexigenic hormone in mealtime hunger and meal initiation.  We have since focused in rats and humans on mechanisms mediating mealtime ghrelin fluxes.  We find that the preprandial ghrelin surge is entrainable to the timing of habitual meals and is triggered by the sympathetic nervous system.  During meals, ghrelin levels are suppressed dose-dependently by the number of calories ingested.  For a given number of calories, proteins suppress ghrelin levels the most, and lipids the least.  Carbohydrate ingestion initially suppresses ghrelin, followed by a rebound to above the pre-meal baseline.  These findings have implications for some of the mechanisms by which popular diets influence appetite.

Ghrelin levels are not affected by gastric, duodenal, or jejunal distention, nor by direct contact between enteral nutrients and gastric or duodenal ghrelin cells.  Meal-related ghrelin suppression results initially from increased intestinal osmolarity, with later contributions from postprandial insulin, both of which are minimal following fat ingestion.  Part of meal-related ghrelin suppression is mediated by the enteric nervous system, but not the vagus nerve.  Gastrointestinal neural transmission involving both nicotinic cholinergic receptors and 5-hydroxytriptamine-3 receptors appears to be involved in the relevant enteric nervous signaling. 

Ghrelin levels respond adaptively to changes in body weight from various interventions, increasing in proportion to weight loss and decreasing with weight gain, consistent with a role in long-term body-weight regulation.  However, humans who have undergone gastric bypass surgery (but not gastric banding) have reduced 24-h ghrelin profiles, which may contribute to weight loss.  Humans with Prader-Willi syndrome have markedly increased ghrelin levels, possibly contributing to their hyperphagic obesity.  This elevation is not observed during the first few years of life, i.e., before hyperphagia and obesity develop.  Ghrelin levels are also elevated in restrained eaters, including among those who are at their lifetime maximal weight. 

Interactions of ghrelin and anorexigenic hormones in hypothalamic intracellular signaling.  Ghrelin exerts opposite effects on feeding behavior, energy homeostasis, and hypothalamic neuronal activity compared with the anorexigenic hormones insulin and leptin.  Hence, we have tested whether ghrelin antagonizes hypothalamic intracellular signaling events triggered by insulin and leptin, and vice versa.  Our studies show that modulation of hypothalamic PI3K signaling is not required for ghrelin’s orexigenic effects, and that insulin and ghrelin do not oppose one another's actions in the hypothalamus through this pathway.  In contrast, intracellular signaling events triggered by ghrelin and insulin within the hypothalamus converge competitively along PKA pathways, and PKA activation is necessary for ghrelin’s full orexigenic effects. 

Effects of Ghrelin on Reward-Related Feeding Behaviors 

We have shown that ghrelin acts at the hypothalamus and hindbrain, sites that regulate homeostatic feeding behavior.  In collaboration with Allen Levine, we helped show that ghrelin also increases food intake by acting on mesolimbic reward sites – the ventral tegmental area and nucleus accumbens.  Reward feeding involves processes that affect the "liking" of food (palatability), in which endogenous opioids are primary mediators, and those that affect the "wanting" of food (motivation), in which dopamine is a primary mediator.  We seek to determine which of these primarily explains ghrelin’s orexigenic effects.  We find that ghrelin does not increase food palatability, as measured using lickometry, and pharmacologic blockade of mesolimbic opioid signaling does not affect feeding induced by VTA ghrelin injections.  In contrast, working with Dianne Lattemann we found that ghrelin administration markedly increases the work (bar pressing) that animals will perform to obtain food, and blockade of brain dopamine signaling attenuates ghrelin-induced feeding.  Our results indicate that although ghrelin does not increase food palatability, it does increase animals’ motivation to obtain food.

Roles of Ghrelin in Meal Initiation 

Our data implicate ghrelin in mealtime hunger and meal initiation, in part because circulating levels surge before meals.  To clarify the physiologic significance of this, we seek to determine whether genetic or pharmacologic ablation of ghrelin signaling impairs meal initiation and food intake under various feeding conditions.  In ghrelin-receptor knockout mice, we have studied the details of meal initiation, meal termination, the microstructure of eating (lickometry), overall food intake, and body weight.  Because of initial negative results using a mixed genetic background, we have back-crossed the mutants for 10 generations against C57BL/6 and plan to use those animals to complete these studies.

Mechanisms by Which Ghrelin Suppresses Insulin Secretion 

We and others have found that ghrelin administration suppresses insulin secretion in vivo.  In collaboration with Ian Sweet, we have investigated the mechanisms for this action in isolated, perifused rat islets.  We find that ghrelin suppresses insulin secretion by mechanisms analogous but opposite to those of the incretins, GLP-1 and GIP.  These observations suggest that ghrelin acts as an “anti-incretin” hormone.

 
Characterize the Effects of Multiple Cycles of Weight Loss and Regain on the Body-Weight Regulatory System

We are characterizing the effects on energy homeostasis of repeated cycles of weight loss and regain.  We serially subject rats to 50% caloric restriction until they lose 20% of body weight, followed by ad libitum feeding until they regain a stable weight.  Relevant hormones (e.g., ghrelin, leptin, insulin) are measured throughout.  We find that following each cycle of weight loss and recovery, weight-cycled rats return to a stable body weight that is progressively lower than that of ad-lib-fed controls.  The time taken to reach a 20% weight loss, and the time to recover a stable body weight, both increase with each successive dieting cycle.  When rats that had been subjected to serial weight-loss/regain cycles for a year were subsequently fed ad libitum for many months, they durably defended lower body weights and adiposity (especially visceral fat), with better glucose tolerance, compared with rats never subjected to caloric restriction.

Our data indicate that repeated bouts of diet-induced weight loss followed by weight recovery confers long-lasting metabolic benefits, causing rats to defend a lower level of body weight and adiposity, with improved glucose homeostasis, even long after dieting ceases.  We hypothesize that weight-cycled rats are spared some of the toxic effects of nutrient excess compared with continuously ad-lib-fed rats, rendering them more sensitive to leptin and/or insulin.  We plan to examine this hypothesis in future experiments, starting with a study of hypothalamic neuropeptide expression in previously weight-cycled vs. control rats.  These findings have clinical implications regarding the utility of repeated dietary weight-loss attempts, which may be beneficial in the long run, even if they are not individually successful in the short run.

Representative Publications:

Cummings DE, Brandon EP, Planas J, Idzerda R, McKnight GS.  Genetically lean mice derived by targeted disruption of the RII subunit of protein kinase A.  Nature 382:622-626 (1996).

Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger EP, Purnell JQ.  Human plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery.  New Engl J Med 346:1623-1630 (2002).

Overduin J, Frayo RS, Grill HJ, Kaplan JM, Cummings DE.  Role of the duodenum and macronutrient type in ghrelin regulation.  Endocrinology 146:845-850 (2005).

Flum DR, Devlin A, Wright AS, Figueredo E, Alyea E, Hanley PW, Lucas MK, Cummings DE.  Development of a porcine Roux-en-Y gastric bypass survival model for the study of post-surgical physiology.  Obes Surg 17:1332-1339 (2007). 

Foster-Schubert KE, Overduin J, Prudom CE, Liu J, Callahan HS, Gaylinn BD, Thorner MO, Cummings DE.  Acyl and total ghrelin are suppressed strongly by ingested proteins, weakly by lipids, and biphasically by carbohydrates.  J Clin Endocrinol Metab 93:1971-1979 (2008).

View Expanded Publication List

Current Collaborations:

Within the Diabetes and Obesity Center of Excellence, The Diabetes and Endocrinology Research Center, and Their Affiliated Members
Denis Baskin, PhD
Ernie Blevins, PhD
Dianne Figlewicz-Lattemann, PhD
David Flum, MD, MPH
Steven Kahn, MB, ChB
Bob Knopp, MD
Mario Kratz, PhD, MSc
Anne McTiernan, MD, PhD
Greg Morton, PhD
Michael Schwartz, MD
Ian Sweet, PhD
Jay Taborsky, PhD
Michi Yukawa, MD, MPH

Lab Members:

Scott Frayo, Lab Manager
Ellen Schur, MD.  Dr. Cummings is the primary mentor on her NIH K23 award.
Karen Foster-Schubert, MD.  Dr. Cummings is the primary mentor on her NIH K12 award.