TwitterRSSFacebookDirectors Blog
  Disorders A - Z:   A    B   C    D    E    F    G    H    I    J    K    L    M    N    O    P    Q    R    S    T    U    V    W    X    Y    Z

You Are Here: Home  »  News From NINDS  »  Proceedings  » 

Skip secondary menu

Workshop: The CNS and Glycemic Control

Image of the CNS and Glycemic Control Poster
September 14–15, 2009
Bethesda Hyatt Hotel

Workshop: The CNS and Glycemic Control Participant List

Executive Summary

The CNS plays vital roles in homeostatic regulation of systemic glucose levels. CNS mechanisms in glycemic control have been demonstrated in basic studies of the hypothalamus, clinical studies on iatrogenic hypoglycemia in Type 1 and Type 2 Diabetes, and glucose management in the intensive care setting. The concept for this workshop originated from multiple and ongoing discussions among officials at NINDS, NIDDK, NHLBI, NICHD, The Department of Defense, and The Juvenile Diabetes Research Foundation. This workshop was organized and charged with reviewing current knowledge in the pertinent, but largely separate research areas of glycemic control and neuroscience to identify: (1) research opportunities, (2) key needs of the relevant research communities, and (3) possible collaborations between diabetes and neuroscience researchers including those working on physiological phenomena that are sensitive to circadian rhythms and sleep. Leading clinical and basic researchers in the fields of diabetes and neuroscience met for two days. The following were major outcomes:

Training and Research Opportunities


Develop K or T program opportunities for creation of interdisciplinary training of scientists to work at the interface of neuroscience and metabolism.

Promote a symposium at a major national meeting (e.g., Society for Neuroscience) focusing on CNS regulation of systemic metabolism, recruiting presentations from major figures in both metabolic and neural fields

Promote a special issue of a significant neuroscience journal on the CNS and systemic metabolic issues and diabetes


  1. Determine the molecular mechanisms responsible for brain glucose sensing in order to develop drugs to counteract defective hypoglycemic counterregulation in diabetes.
  2. Determine the neural or metabolic mechanisms by which circadian alterations, or interrupted or loss of sleep leads to impaired glucose control, and whether these processes differ in children vs. adults.
  3. Determine how the brain adapts to recurrent hypoglycemia and compare this to the adaptation to intermittent ischemia (preconditioning). This could be used to develop therapeutics to protect the brain from injury.
  4. What is the mechanism for brain injury during acute hypoglycemia?   The role of subclinical seizures and post-hypoglycemia hyperglycemia mediated free radical injury.
  5. Determine the mechanisms by which gut signaling alters appetitive drives and counters glycemic changes.
  6. Determine whether neuropathy induced by glycemic processes interferes with efferent vagal or spinal axons influencing pancreatic and liver action, or with afferent nerves underlying gut signaling to alter glucose regulation.


  1. Determine the neural or metabolic mechanisms by which interrupted or loss of sleep leads to impaired glucose control, and whether these processes differ in children vs adults.
  2. Determine the processes by which circadian mechanisms exert influences on insulin sensitivity and glucose regulation.
  3. Determine mechanisms underlying co-morbidity of sleep disordered breathing, obesity, and T2 diabetes.
  4. Determine the mechanisms by which central neural injury results from hyperglycemia, the processes by which this injury is enhanced, including potential interactions with sleep-disordered breathing, and interventions to ameliorate injury.
  5. Determine the mechanisms by which gut signaling influences glucose control, and how bariatric surgery may modify that regulation.
  6. Determine the interactions between neuropathy of efferent and afferent nerves and influences of gut signaling on glucose regulation.
  7. Determine central and peripheral processes by which cerebrovascular regulation can be impaired in T2 diabetes, and what interventions will minimize consequences.

Agenda and Presentation Summaries

Monday, September 14, 2009

7:30–8:30 a.m.Registration and Continental Breakfast
8:30–8:40 a.m.Welcoming Remarks, Story Landis, Director, NINDS
8:40–8:50 a.m.Welcoming Remarks, Gregory Germino, Deputy Director, NIDDK
8:50–9:00 a.m.Introduction and Orientation, Merrill Mitler, NINDS

The concept for this workshop originated from discussions with officials at NIDDK, NHLBI, NICHD, The Department of Defense, and The Juvenile Diabetes Research Foundation. All parties recognize the importance of the CNS in homeostatic regulation of systemic glucose levels. CNS mechanisms in glycemic control have been demonstrated in basic studies of the hypothalamus, clinical studies on iatrogenic hypoglycemia in Type 1 and Type 2 Diabetes, and glucose management in the intensive care setting. Also realized was the potential for elevated or reduced glucose levels to induce neural injury which has the potential to exert deleterious effects on cognitive, emotional, and autonomic nervous system control, and which may exacerbate the underlying glucose regulation deficits. The advent of continuous glucose monitoring technology is a new development that greatly facilitates 24-hour studies of glycemic control and permits novel interdisciplinary research on the CNS and glucose in animal and human subjects. Continuous glucose data from diabetic and critically ill patients show that derangements in glucose levels occur disproportionately during the hours of sleep. Reasons for the relationship between sleep and abnormal glucose levels, while not fully known, do indicate that neural injury accompanying impaired glucose control may share features of cellular damage associated with ischemia during sleep. The objectives of this workshop are to review current knowledge in these disparate, but inter-related, research areas and identify: (1) research opportunities, (2) key needs of the relevant research communities, and (3) development of collaborations between diabetes and neuroscience researchers including those working on physiological phenomena that are sensitive to circadian rhythms and sleep.


9:00–9:20 a.m.Hypoglycemia and Defective Counterregulation: The Factor Limiting the Benefits of Intensive Insulin Treatment, Robert Sherwin

Large multicenter clinical trials in Type 1 diabetic patients have shown substantial long-term benefits of tight glycemic control. However, iatrogenic hypoglycemia and defective counterregulatory responses to hypoglycemia (e.g. the release of epinephrine and glucagon) have emerged as major limitations to the intensive glucose control of diabetes and are causes of CNS injury. Animal studies of hypoglycemia have demonstrated several noteworthy phenomena. Responses to counteract hypoglycemia appear to adapt by showing reduced responses to successive hypoglycemic challenges. There is a faster recovery, if such hypoglycemic episodes are eliminated. Responses to hypoglycemia are also reduced during sleep, which may lead to severe hypoglycemic events when corrective action is more difficult. Key glucose sensing neurons reside in the ventromedial hypothalamus. They are glucose excitatory neurons and glucose inhibitory neurons. One of the transmitter systems implicated in activation of the CNS response to hypoglycemia and its adaptation to recurrent antecedent hypoglycemia is GABA. The brain’s adaptation to hypoglycemic episodes may be a two-edged sword. Its adaption may be protective in the service of becoming more efficient. However, this adaptation may also reduce the brain’s ability to detect and respond to glucose deficit. Answers to this serious clinical problem caused by altered brain signaling and metabolism is but one example of why there is a need for closer interactions between diabetologists and neuroscientists. Other major health problems at the interface of diabetes and neuroscience include obesity, the metabolic syndrome and Alzheimer’s disease.

9:20–9:30 a.m.Discussion
9:30–9:50 a.m.Diabetes, Hyperglycemia, and Stroke, Bruce Ransom

While diabetes has many systemic effects that raise the risk of stroke, including promoting atherosclerosis, it also routinely causes hyperglycemia which seems to be an independent risk factor correlated with stroke severity. Animal studies, primarily on rodents, generally show that when hyperglycemia is present just before the onset of brain ischemia, tissue injury is increased compared to control animals with normal serum glucose. Ischemia-induced acidosis is greater with hyperglycemia and seems to directly enhance tissue injury. Recent clinical studies, however, suggest that strokes, which primarily affect white matter (i.e., nonsynaptic areas of the brain consisting exclusively of axonal tracts), called ‘lacunar’ strokes, actually have better clinical outcomes in the presence of higher serum glucose concentrations. More preclinical research on the subject of white matter ischemic injury and hyperglycemia is needed.

9:50–10:00 a.m.Discussion
10:00–10:20 a.m.Circadian Disruption and Metabolic Disease, Joseph Bass

The genetic regulators of circadian rhythms play complex roles in metabolic change throughout the 24-hour day. Life forms on the earth’s surface were shaped by environmental changes associated with the earth’s rotation. Among the strongest of these rotation-related changes is the availability of energy. Darwin recognized the influences of the earth’s 24-hour day; he and his son took an interest in circadian biology. Seymour Benzer applied forward genetics to the neuroscience of locomotor activity in the fly. We have applied forward genetics in an analogous way to pathways regulating energy balance. Ablation of the suprachiasmatic nucleus disorganizes circadian rhythms. The arcuate nucleus is important for energy balance. Fred Turek and Joe Takahashi found in their forward genetics work a mouse with a period length 6 standard deviations longer than normal. Numerous studies by several groups have worked out the genetic activity underlying circadian rhythms and variability in these rhythms. At the core is a transcription/translation feedback loop, expressed within neurons of the SCN and also nearly all peripheral tissues which oscillate according to a 24 hour cycle in nearly all tissues. This feedback loop is affected by numerous other genetic factors. 3 to 10 percent of all genes oscillate with 24-hour periodicity. This pervasive 24-hour pattern is interconnected with metabolism. It obviates our waking up in the middle of the night requiring feeding. Across all organisms, the 24-hour pattern organizes feeding so that in fits into the overall activity-inactivity pattern. Another piece of evidence that feeding and sleep are interconnected is the discovery that the sleep disorder narcolepsy can be caused by disruption in orexin (hypocretin) transmission because orexin is an important neurotransmitter for feeding. When the temporal organization of feeding is disrupted, regulation of body weight is also disrupted. There is clear influence of the biological clock in pancreatic function, the sirtuin pathway and the production of nicotinamide adenine dinucleotide (NAD). In sum, in normal healthy conditions there is circadian synchronizing both at the physiologic and behavioral level between sleep/wakefulness states, feeding and fasting and the energy cycle. But in conditions of disrupted circadian synchrony either through environmental change (e.g. shift work, sleep loss or even diet change), the cycles in peripheral tissues are no longer synchronized with the behavioral cycle and this disruption is one of the etiologic factors for metabolic disease and for the sorts of derangements of glucose levels we see in diabetes.

10:20–10:30 a.m.Discussion
10:30–11:00 a.m.BREAK

INJURY—Charles Wade, Chair

11:00–11:20 a.m.Hypoglycemia and Hypoxia: Similarities and Differences in Patients With Sleep-Disordered Breathing, Ronald Harper

Interrupted sleep or sleep deprivation leads to elevated circulating glucose, and the combination of obesity and diabetes in patients result in an exceptionally high incidence of obstructive sleep apnea. Obstructive sleep apnea is accompanied by major injuries to brain sites which mediate memory, cognitive, and emotion functions, as well as to cerebellar coordination structures and cortical areas that influence hypothalamic action and autonomic outflow. This injury appears as regional loss of gray matter, elevated T2 relaxometry values, loss of axonal integrity, and impaired functional magnetic resonance imaging responses to ventilatory and autonomic challenges. The mechanisms underlying the injuries are unclear, but may include hypoxic processes leading to excitotoxicity, as well as oxidative mechanisms accompanying reoxygenation after apnea. The neural damage is enhanced if patients also have Type 2 diabetes. In developmental sleep-disordered breathing conditions associated with hypoglycemia, injury appears in brain areas that mediate self-care and reckless behavior, an outcome which appears to be shared in children with Type 1 diabetes with respect to self-maintenance of glucose levels.

11:20–11:30 a.m.Discussion
11:30–11:50 a.m.Hypoglycemic Brain Injury and the Impact of Glucose Delivery, Raymond Swanson

Hypoglycemic neuronal death is not simply a consequence of energy failure, but results instead from a process involving excitotoxicity and oxidative stress. This process is triggered by glucose reperfusion after the hypoglycemic period, and is exacerbated by rapid glucose correction. Glucose reperfusion induces oxidative stress by fueling superoxide production by neuronal NADPH oxidase. This process also contributes to the deleterious effect of hyperglycemia in stroke, and may also contribute to deleterious effects of hyperglycemia in other clinical settings.

11:50–12:00 a.m.Discussion
12:00–12:20 p.m.Processes in CNS Injury in Intermittent Hypoxia, David Gozal

Sleep disordered breathing (SDB) and obesity co-segregate in the pediatric population and operate as mutual risk factors; both these conditions increase the risk for insulin resistance and type 2 diabetes. While the prevalence of SDB among diabetic patients has not been definitively assessed, preliminary data indicate a high frequency of SDB among young diabetic patients. SDB is characterized by intermittent hypoxia (IH) and sleep fragmentation (SF). Studies in children and adults would indicate that both of these conditions independently increase the risk for insulin resistance. Experimental animal models of chronic IH or SF during sleep have been developed and confirm the presence of insulin resistance. Furthermore, IH also leads to increased apoptosis of beta cells in the pancreas. Both IH and SF lead to behavioral and cognitive dysfunction and cardiovascular complications, and the interaction between hyperglycemia and hypoglycemia and SDB could potentiate CNS and CV end-organ morbidity. Taken together, SDB not only increases the risk for diabetes, but also SDB and diabetes may either synergistically or additively enhance the risk for associated and overlapping morbidities.

12:20–12:30 p.m.Discussion
12:30–1:30 p.m.LUNCH

BRAIN FUEL SENSING—Charles Mobbs, Chair

1:30–1:50 p.m.Molecular Mechanisms of Brain Glucose Sensing, Barry Levin

Specialized neurons localized in anatomically discrete locations throughout the brain utilize glucose as a signaling molecule to regulate their activity. These neurons “sense” glucose through the generation of ATP, AMP, reactive oxygen species and nitric oxide and other products of glucose metabolism. Glucokinase acts as a gatekeeper of glucose metabolism in a majority of these neurons, many of which also sense long chain fatty acids, lactate, ketone bodies and hormones such as insulin and leptin. Importantly, such metabolic sensing neurons depend upon close interactions with astrocytes and, in the hypothalamus, with tanycytes lining the third ventricle. Together, they form a distributed network which is involved in regulating the neurohumoral responses to hypoglycemia and energy homeostasis, and provide a potential therapeutic target for correcting abnormalities of these processes in diabetes and obesity.

1:50–2:00 p.m.Discussion
2:00–2:20 p.m.Central Versus Peripheral Glucose Sensing and Response to Hypoglycemia–Alan Cherrington

Activation of hypoglycemic counterregulatory responses requires effective detection of a falling glucose level, before it reaches dangerous levels. A complex network of glucose sensors that is present in both the CNS and in extracerebral tissues serves this purpose. In non-diabetic subjects, hypoglycemia provokes a multi-tiered defense system that includes suppression of endogenous insulin secretion and the secretion of the counterregulatory hormones, glucagon and epinephrine. In addition, hypoglycemia per se at the liver increases glucose production, whereas it does not affect lipolysis. Data were presented that suggest insulin-induced hypoglycemia may increase plasma glucagon through a CNS effect. Moreover, insulin-induced hypoglycemia increases the sympatheoadrenal response through a CNS effect. However, the rate of fall of glucose levels may exert an influence on the site at which glucose sensing occurs. Studies in rodents suggest that both central and peripheral (portal vein) glucose sensors drive the sympathoadrenal response, with the latter becoming more important when the rate of fall of glucose is slow.

2:20–2:30 p.m.Discussion
2:30–2:50 p.m.Hypothalamic Energy Sensing and the Role of Lipids, Gabriele Ronnett

Although usually known for their role in energy storage, fatty acids may also serve as sensors of organismal energy status. Modulation of fatty acid metabolic pathway enyzmes, specifically fatty acid synthase, carnitine palmitoyltransferase-1, and glycerol-3-phosphate acyltransferase, result in profound weight loss, decreased food intake, and increased peripheral energy utilization. AMP-activated kinase appears to mediate a large portion of these effects. Pathophysiological alterations in energy balance, as seen in stroke, over-activate AMPK; in this scenario, modulators of fatty acid metabolism provide neuroprotection in animal models.

2:50–3:00 p.m.Discussion
3:00–3:30 p.m.BREAK

HOMEOSTASIS—Ronald Harper, Chair

3:30–3:50 p.m.Interactions of Wake-Sleep and Circadian Systems With Feeding, Clifford Saper

Recent work in humans suggests that sleep restriction may cause obesity and metabolic syndrome. However, long-term trials of sleep restriction, or intervention, are lacking at this time. In animals, all studies so far have been short-term, and involve manipulating animals in ways that are stressful. We used lesions of the ventrolateral preoptic area to cause long term (60 days) reduction in sleep by 30-35%. These studies were consistent with earlier animal literature, showing that sleep restriction itself appears to cause reduced weight gain, glucose, insulin, lipids and cholesterol. We are currently examining the effects of a high fat diet combined with sleep loss.

3:50–4:00 p.m.Discussion
4:00–4:20 p.m.Circadian Rhythms and Sleep: Influence on Glucose Homeostasis, Eve Van Cauter

Circadian rhythms and sleep modify glucose tolerance, with decreased tolerance during sleep, and during evening hours over the morning. Insulin exerts hypnotic effects and promotes GABAergic activity. Sleep deprivation impairs glucose tolerance substantially; one mechanism may result from the 7-8 % reduced glucose utilization in the sleep-deprived brain, modifying counterregulatory processes. Sleep debt significantly decreases leptin release, with the inverse leptin and cortisol relationship possibly promoting insulin resistance; a phase shift occurs in the leptin-cortisol cycle following sleep deprivation. Disrupting sleep, to mimic aging-related slow wave sleep changes, greatly reduces insulin sensitivity, and sleep-disrupted waking times increases fasting glucose significantly (30 min, 15 mg/dl; 2 hrs, 50 mg/dl). Obstructive sleep apnea occurs in a large proportion (up to 86%) of diabetic patients who are obese; even mild OSA elevates HbA1c.

4:20–4:30 p.m.Discussion
4:30–4:50 p.m.Gut and Adipocyte Hormones: Effects on the Brain and Glucose Metabolism, Randy Seeley

GLP-1 is a gut peptide made in both the ileum and discrete regions of the brainstem involved in the regulation of glucose. While the predominant model is that gut derived GLP-1 acts as a hormone on key organs involved in glucose homeostasis, increasing evidence points to other actions of GLP-1. GLP-1 appears to act both in the CNS and peripheral nerves to regulate glucose homeostasis.

4:50–5:00 p.m.Discussion

Working Dinner

6:00–9:00 p.m.Prospects for Closed Loop (Artificial Pancreas) Systems in Diabetes, William Tamborlane

The introduction of external real-time continuous glucose monitoring systems in combination with external insulin pumps has made the development of a practically applicable closed-loop artificial pancreas possible. Short-term, clinical research center studies have demonstrated the feasibility of this approach, particularly in developing children. While there remain a number of obstacles to overcome, and clinical strategies to improve system performance need to be tested, the prospects for a mechanical solution for optimizing glucose control in diabetes are brighter than ever.


Tuesday, September 15, 2009


8:30–8:50 a.m.Influence of Hyperglycemia on Autonomic Neuropathy, Eva Feldman

The Epidemiology of Diabetes Interventions and Complications (EDIC) study, a prospective observational follow-up of the Diabetes Control and Complications Trial (DCCT) cohort, reported persistent benefit of prior intensive therapy on retinopathy and nephropathy in type 1 diabetes mellitus. We evaluated the effects of prior intensive insulin therapy on the prevalence and incidence of cardiac autonomic neuropathy (CAN) in former DCCT intensive and conventional therapy subjects 13 to 14 years after DCCT closeout. DCCT autonomic measures (R-R variation with paced breathing, Valsalva ratio, postural blood pressure changes, and autonomic symptoms) were repeated in 1226 EDIC subjects in EDIC year 13/14. Logistic regression models were used to calculate the odds of incident CAN by DCCT treatment group after adjustment for DCCT baseline covariates, duration in the DCCT, and quantitative autonomic measures at DCCT closeout. In EDIC year 13/14, the prevalence of CAN using the DCCT composite definition was significantly lower in the former intensive group versus the former conventional group (28.9% versus 35.2%; P=0.018). Adjusted R-R variation was significantly greater in the former DCCT intensive versus the former conventional group (29.9 versus 25.9; P<0.001). Prior DCCT intensive therapy reduced the risks of incident CAN by 31% (odds ratio, 0.69; 95% confidence interval, 0.51 to 0.93) and of incident abnormal R-R variation by 30% (odds ratio, 0.70; 95% confidence interval, 0.51 to 0.96) in EDIC year 13/14. The data demonstrate that although CAN prevalence increased in both groups, the incidence was significantly lower in the former intensive group compared with the former conventional group. The benefits of former intensive therapy extend to measures of CAN up to 14 years after DCCT closeout.

8:50–9:00 a.m.Discussion
9:00–9:20 a.m.Early and Subtle Changes in Cerebral Autoregulation in Diabetic Patients, Max Hilz

The high incidence of stroke in diabetes mandates examination of cerebral autoregulation (CA), essential for maintenance of constant blood supply to the brain during momentary changes in blood pressure, and, when dysfunctional, a major contributor to ischemic and hemorrhagic stroke. Regulation of CA depends on multiple influences, several of which can be injured or altered in diabetes, and includes sympathetic and parasympathetic innervation to the cerebral arteries. That innervation can be modified by central aminergic vasoconstor pathways; large serotonergic and other aminergic nerves accompany some of the arteries, e.g. the basilar artery, extraparenchymal arteries are innervated mainly from the trigeminal ganglion. Endothelial processes that modify vasoconstriction can be damaged in diabetes and in conditions, e.g., obstructive sleep apnea, heart failure, that often accompanying diabetes. Those conditions can also modify carbon dioxide and oxygen levels, and thus, CA. Among significant factors affecting vasoconstriction and dilation, autacoids, e.g., NO, are significant; static and dynamic myogenic factors (Bayliss component) are important considerations. Cerebral autoregulation can be assessed by examining phase lags to imposed stimulation of the carotid bodies by 0.1 Hz neck suction, and determining the transfer function phase angle between CBFV and blood pressure oscillations from that stimulation. In clinical environments, 6 cycle/min metronomic breathing can be used as a blood pressure challenge to show mild impairments in CA, even in patients with concomitant autonomic neuropathy.

9:20–9:30 a.m.Discussion
9:30–9:50 a.m.Hypoglycemic Seizures, Peter Carlen

Hypoglycemic seizures are relatively little studied, but play a large role in the side effects of insulin therapy, especially in Type I diabetes in developing children. In vitro data, supporting the hypothesis that hypoglycemic seizures, during limited substrate availability (i.e. glucose), exacerbate neuronal damage, both functionally and histologically, were presented. At the cellular level, hypoglycemia caused depolarization of several different neuronal types, with little correlation between the seizure-like activity and the neuronal activity, raising the question as to which cells underlie seizure electrogenesis. Glia were hypothesized to play a major, and heretofore not investigated role in seizure generation and in neuroprotection. Preliminary hypotheses and data were presented showing an important role for gap junctions and possibly pannexin hemichannels in generation of hypoglycemic seizures and in neuroprotection during these seizures. Finally, parallels of the similarity of SUDEP (sudden unexpected death in epilepsy) and the “Dead in Bed” syndrome in diabetes, and their pathogenesis possibly being attributable to brainstem seizure activity with cardiorespiratory nuclei dysfunction were presented.

9:50–10:00 a.m. Discussion
10:00–10:30 a.m.BREAK


10:30–10:50 a.m.Glycemic Control, Brain Structure, and Cognition, Tamara Hershey

Exposure to glycemic extremes during development has detectable and negative effects on human brain structure and function in T1DM. Hyperglycemia and severe hypoglycemia each exert unique effects; hyperglycemia is associated with reduced verbal IQ and reduced gray and white matter volumes in the precuneus/cuneus, whereas severe hypoglycemia is accompanied by a decline in delayed-memory performance and reduced gray matter volume in the left temporal-parietal-occipital cortex, as well as enlarged hippocampi. Further prospective analyses are needed to determine how this exposure affects the developmental trajectory of the brain, whether these effects resolve or are increased over time, and how they relate to cognitive outcome.

10:50–11:00 a.m.Discussion
11:00–11:20 a.m. Neuroimaging for Assessment of Brain Function in Diabetes, Stephanie Amiel

Regional changes in either glucose uptake, content and metabolism (glucose positron emission tomography) or in brain perfusion (water positron emission tomography or the evolving techniques in functional magnetic resonance imaging of arterial spin labeling) or oxygenation status (Blood Oxygen Level Dependent fMRI) provide surrogate markers of neuronal activation in response to, or during hypoglycemia. We showed that, in addition to activation of the hypothalamic-pituitary-adrenal access, brain regions involved in interoception and appetite control are activated, and regions involved in hedonic perception and reward may be de-activated, together with areas involved in memory, vision and coordination. Water PET allowed examination of the evolution of these responses over time and into recovery. Differences in regional brain responses to hypoglycemia in diabetic patients with hypoglycemia unawareness and the resultant high risk of severe hypoglycemia include reduced activation of stress response centers and reversal of the deactivation of such regions as parts of the orbitofrontal cortex, patterns consistent with stress-desensitization. These data suggest that such patients will require more than just education about hypoglycemia avoidance to adjust their therapy to restore awareness, since they receive no internal cues to the danger of each episode of severe hypoglycemia. The problem is a particular concern in developing children, since the unawareness may be mistaken for youthful indifference.

11:20–11:30 a.m.Discussion
11:30–11:50 a.m.Adaptations of Brain Metabolism to Hypoglycemia and Hyperglycemia: MRS Studies, Elizabeth Seaquist

The human brain is vulnerable to effects of hypoglycemia and hyperglycemia. In patients with Type 1 diabetes and hypoglycemia unawareness, steady-state brain glucose concentrations are higher than matched controls, apparently because of an up-regulation of glucose transport across the blood brain barrier. This up-regulation of glucose transport following hypoglycemia may represent one mechanism contributing to hypoglycemia unawareness. Another potential mechanism may be supercompensation of brain glycogen stores following hypoglycemia. While the impact of diabetes on brain glycogen metabolism remains unknown, we found that brain glycogen is mobilized during periods of modest hypoglycemia, and that brain glycogen content is greater following hypoglycemia than immediately before.

11:50–12:00 a.m.Discussion
12:00–1:00 p.m.LUNCH


1:00–1:20 p.m.Therapeutic Interventions Derived From Animal Studies, Rory McCrimmon

Individuals with Type 1 diabetes suffer repeated hypoglycemia as a consequence of their insulin therapy and of widespread defects in the normal neuroendocrine response to hypoglycemia. Key to understanding the impaired glucose counterregulatory response will be unraveling of CNS glucose-sensing mechanisms over the course of development. In addition, the complex interactions between altered central insulin signaling in diabetes, glucose-sensing and the stress response needs to be explored. Parallels between the central adaptation to repeated hypoglycemia (hypoglycemic pre-conditioning or tolerance) and the central response to other stressors (e.g. hypoxia) suggest that much can be gained through the closer interaction between physicians and scientists working in the metabolic and neurological fields. Finally, to develop therapies designed to prevent or minimize severe hypoglycemia in insulin-treated diabetes will require pre-clinical testing in physiologically relevant rodent models of diabetes.

1:20–1:30 p.m.Discussion
1:30–1:50 p.m.Therapeutic Interventions for Defective Glucose Counterregulation in Human T1DM, Philip Cryer

Interventions could include 1) normalization of attenuated sympathoadrenal activation/actions, 2) restoration of absent increments in glucagon secretion/actions, or 3) restoration of absent decrements in insulin secretion/actions. As little as 2-3 weeks of scrupulous avoidance of hypoglycemia reverses hypoglycemia unawareness and improves the attenuated epinephrine component of defective glucose counterregulation, but this is difficult to accomplish and maintain. Oral selective serotonin reuptake inhibitors (fluoxetine, sertraline), an intravenous opioid antagonist (naloxone) and fructose infusion increase plasma epinephrine responses to hypoglycemia. An oral β2-adrenergic agonist (terbutaline) prevents hypoglycemia. Adenosine antagonists (caffeine, theophylline) and a drug that decreases GABA levels (modafanil) increase symptoms of hypoglycemia. Subcutaneous glucagon and oral amino acids (which increase circulating glucagon) prevent hypoglycemia. None of these have been subjected to randomized controlled trials. Even partial plasma glucose regulated insulin secretion, following successful islet transplantation, can produce a normal HbA1C with no hypoglycemia. When they become practical, new treatment methods that provide plasma glucose regulated insulin secretion (e.g., transplantation) or replacement (e.g., closed-loop insulin replacement) will likely eliminate hypoglycemia from the lives of people with diabetes.

1:50–2:00 p.m.Discussion
2:00–2:30 p.m.BREAK


2:30–4:30 p.m.Co-Chairs, Ronald Harper and Robert Sherwin

• Gaps in Knowledge

• Research Opportunities

• Translational Opportunities

DR. HARPER: There was much presented in the last two days that demonstrates the potential for areas within neuroscience to directly benefit the metabolic disorders that are of concern here. Of particular interest was the evidence that other disturbed physiological conditions, such as altered sleep, or pathologic conditions within sleep significantly impact glucose regulation. I was very much impressed with the evidence that Dr. Cauter and others presented showing the extraordinary derangement mild sleep deprivation produces in overall regulation of glucose, and how circadian effects modify these phenomena. Dr. Gozal documented some of the consequences of disordered breathing during sleep in development, and the resulting insulin resistance in children, as well as the significant cognitive consequences when that regulation goes awry. These interactions that involve integration of physiological systems across different fields provide an opportunity for insights into glucose regulation that should bear good fruit for the NIH institutes involved. We devoted only limited time on hormonal regulation and circadian processes, but clearly we can’t think about circadian rhythms without considering regulation of such processes, such as melatonin, or temperature rhythms on metabolic processes.

Important also is the effect of disordered breathing during sleep on areas that regulate brain structures directly involved in autonomic regulation and control of hormones involved in diabetes. A number of studies have documented the high incidence of sleep disordered breathing in diabetic patients. There are relationships here that should be helpful for relating neural injury developing from one pathology, sleep disordered breathing, on another, metabolic regulation.

We have heard several descriptions of consequences to the CNS following hypoglycemia and hyperglycemia, with the effects on the brain being largely deleterious, although there was an interesting discussion of how some aspects of compensation may be beneficial. What we heard from Dr. Hershey demonstrates that cognitive damage that can result from both extremes of glycemic control, while Dr. Amiel outlined an aspect of major concern in developing children; namely, the brain modifications that lead to hypoglycemic unawareness, or loss of internal cues, and thus, to even more extreme hypoglycemia. Such loss of self-care cues is found in other developmental conditions associated with injury to affective brain regions, and a focus on neuroprotective interventions is required, not just, as Dr. Amiel points out, educational efforts.

The regulation of glucose as observed in the impressive MRS studies of Dr. Seaquist draw attention to both the regional and overall changes in brain, as well as consequences to the peripheral nervous system associated with Type 1 and Type 2 diabetes. Dr. Feldman outlined some of the consequences of neuropathy accompanying both forms of diabetes, and drew attention to issues of assessment of injury through cardiac variability, and interventions for damage. Attention also should be directed to how neuropathy of fibers to the major structures in the gut that regulate glucose are modified during the course of diabetes. Currently, we rarely speak of disrupted autonomic regulation in terms other than peripheral limb pain, sensory loss, or reduced heart rate variability accompanying neuropathy; however, injury to vagal efferents will also affect pancreatic function, and damage to afferents from the gut will modify signaling of GLP and other gut signals to the CNS. It is apparent from Dr. Seeley’s discussion that gut hormones exert significant influences on glucose control. Gut regulation of hormones and the processing involved in signaling to and from the brain may be heavily impacted by glucose-related injury to those nerves.

Finally, the various mechanisms underlying injury to the brain and peripheral nerves are unresolved. Ray Swanson’s discussion of reperfusion of glucose and the resulting excitotoxicity and oxidative stress processes accompanying the reperfusion provides important insights. There may be significant parallels for what we in the sleep field are considering with intermittent hypoxia and the reoxygenation that accompanies restoration of breathing. It may not be the intermittent hypoxia that is inflicting the injury so much as the reoxygenation following apnea that may be damaging. And, with any of these processes of intermittent hypoxia, substantial changes in vascular regulation emerge partially through respiratory coupling with the cardiovascular system, and partially through the sequence of hypoxic stimulation of the autonomic nervous system, all of which have the potential for inducing injury. We heard a description this morning by Dr. Hilz of how regulatory control of the cerebrovascular system goes out-of-phase in obstructive sleep apnea patients. Those time lags for neural signaling to circulatory processes have the potential to elicit serious injury from excitotoxic injury when cellular needs must wait for adequate perfusion, a special concern when cells are over-driven by seizure discharge, as Dr. Carlen described. We heard about the potential for stroke in diabetes; presumably, ischemic stroke forms part of that concern, and the injury to brain structures which regulate autonomic control of the cerebrovascular may compromise perfusion and contribute to stroke mechanisms. The increase in thrombolytic processes accompanying increased inspiratory loading, the respiratory action mimicking obstructive sleep apnea, may also be playing a role in diabetics with OSA as well.

Thus, there are many interacting issues associated with examination of autonomic control and sleep regulation that we must consider. The CNS is at the heart of both these processes, and we must pull our fields together. Bob and I have been discussing mechanisms by which we can integrate young people from the neuroscience and metabolic disease fields and the processes that might be initiated for this integration..

DR. SHERWIN: The first thing that impressed me is the complexity of the subject matter we have been hearing about. This field is complex at all levels -- the base of the brain, the limbic system and the cortex. These systems are regulated by neurotransmitters and by their substrates. When I grew up, people told me never to look at the brain because it didn’t respond to insulin and really wasn’t regulated by glucose and consequently this is not an area to study. I think most neuroscientists look at diabetes and metabolism from the opposite perspective. For example, they see Alzheimer’s and stroke, but think of diabetes and metabolism as areas that one should not look at. They are too complex, and they really haven’t been trained to study those areas.

We have investigators in two fields; both groups are trained not to examine each other’s field because of complexity. It is better to focus on the areas in which one has been trained. Well, one goal of this meeting is to try to think about how we can change that attitude. There are fundamental questions relevant to not only metabolism and diabetes but to neurological disease that are important and can be addressed if these two groups work together and to think together. So, one of the things I think would be important as we go forward is to think about how either JDRF, or the NIH which has more resources but is constrained to some extent by the missions of individual ICs, can begin to train people, perhaps at the T level as well as the K level, to work in a new field, CNS regulation of systemic metabolism, which really tries to bring together both disciplines? Obviously if you develop purely a neuroscience research program or purely an endocrine metabolism research program or a training program, you need to begin to think about those centers where people begin to talk to each other, and perhaps through joint efforts, to try to develop training environments that could allow for training of the new generation of researchers competent to do CNS regulation of systemic metabolism.

We heard today about glucose sensing. It was clear that this is a very complex area. We are really just scratching the surface. We are beginning to think about the molecules that may be involved. But, we haven’t focused adequately on the neurotransmission and the circuitry of how the hypothalamus and other areas sense glycemic state and are interconnected. These are fundamental questions that go beyond diabetes, go beyond glucose regulation and are important to answer. So, I think that understanding how the brain and periphery sense glucose, talk to each other, connect, as well as what molecules are involved and what neurotransmitters are critical in that circuitry is critical and is fundamental to biology and neuroscience and not just diabetes.

It is also important to think about the endocrine system and how it impacts on the brain. We heard that sleep disorders affect insulin action. Circadian rhythms are probably playing a role. But we really don’t understand those processes at all. We have heard about such molecules as leptin, ghrelin, and the hypocretins. It is likely that these molecules are involved in more than feeding. The likelihood is that a lot of these hormones have effects on other behaviors and the limbic system. Because a molecule affects a specific area of brain doesn’t mean that is all it does. I am sure we will find things are far more complex. I think it is going to turn out that insulin has a critical role brain function. Several investigators have observed that the hippocampus is a very insulin responsive structure. It has a lot of Glu 4 and responds to insulin. If you create insulin resistance in animals, you can induce and impair cognitive function.

Also, we must remember that amyloid biology is regulated by insulin and is relevant to such conditions as Alzheimers Disease. So, the likelihood that this beta cell peptide, that clearly works locally, also talks to the brain in many different ways, not just simply via glucagon. Feeding behavior and reward are probably affected. I think it is likely that hormones like these are going to play a critical role in basic functions. Probably a lot of the hormonal regulation involves basic systems because all these systems evolved with the brain trying to maintain homeostasis as best it could and getting as much information as it could from the periphery.

Then there is the question is hypoglycemia itself. What are its effects on the brain over time? What are the effect over time of hyperglycemia? My interest in this began when I realized we were producing a lot of hypoglycemia early in our treatment regimens for diabetes. It was very clear we didn’t know what hypoglycemia would do over time. And, we still don’t know. We know that acute hypoglycemia is bad if you have seizures. We know that it is probably bad if you have it early in life, when the brain is developing. And, we know that people are developing Type 1 diabetes at a younger age than ever before – sometimes as early as age 2 years. The loss of beta cell mass occurs quicker in those children than it does if you develop Type 1 diabetes when you are 20. Not only do youngsters lose beta cell function much quicker; it is also very hard to regulate glucose in these children. So we have a fundamental problem with those children in preventing loss of cognitive function. We know that in the DCCT, crude estimates of cognitive function clearly did show a gross change. Most of these were people who developed the disease later in life. Most of them didn’t have hypoglycemia unawareness. But, we don’t know what is going to happen when they are 50. Most of the loss of cognitive function over time is really an aging process, and we really haven’t looked at the aging process at all from the perspective of our current treatment regimens, which are very different than they were before. Some of the animal models suggest that recurring hypoglycemia may actually lead to a positive adaptation - that the brain is trying to use fuels more effectively and efficiently to protect itself. It would make sense that the body tries to protect itself from recurrent injury, just as it does with ischemic injury.

I think that there are similarities between intermittent ischemia and intermittent hypoglycemia. Both depend on lack of supply of either oxygen or energy, which is not exactly the same but similar. Investigators in these two areas can learn from each other. The neuroscience community is focused on consequences from the standpoint of ischemia. I think that the diabetes community is focused on consequences from the perspective of hypoglycemia. But, we are talking about probably very similar processes, because the brain and the body doesn’t try to create new mechanisms to deal with similar kinds of problems.

The last issue to mention here is chronic hyperglycemia. Injury from free radicals is surely a contributor to loss of cognitive function over time, and probably contributes to why people with Type 2 diabetes have cognitive deficits as well.

Many of us learned today that strokes were very common, particularly white matter strokes, in people with insulin resistance and Type 2 diabetes. This is less clear in people with Type 1, but I don’t think it has been adequately looked at in people at an older age group.

So the bottom line is that these two communities share common interests and need to begin to think about how we can begin to learn from each other. And, how do we begin to train young people to think from a joint diabetes-neuroscience perspective.

DR. MITLER: First, thank you all for coming. Now, I think it would be appropriate to take note that there have been some very patient people in the audience with distinguished careers in the NIH. I think it might be appropriate for those who are willing to introduce themselves and tell us where you are from and why you have remained so interested for all these hours.

4:30 p.m.Workshop adjourns


Workshop: The CNS and Glycemic Control Participant List

Last Modified October 18, 2015