Neurodegenerative diseases other than those associated with cardiovascular sequelae of diabetes, notably Alzheimer’s disease (AD), mild cognitive impairment (MCI) and Parkinson’s disease (PD) disproportionately affect patients with Type 2 Diabetes. Cognition declines more rapidly in the elderly with diabetes and seems to be negatively affected even by pre-diabetic increases in glycaemia. Some pathological processes, particularly amyloidosis, impaired insulin signalling and low-grade inflammation, characterize both conditions. In experimental settings a range of anti-diabetic drugs can modify cognitive performance and neuronal functions. This has led to optimism that targeting classical diabetes mechanisms may improve outcome in AD, MCI and PD and the term type 3 diabetes has accordingly been coined for AD. In patients with diabetes as well as in normoglycaemic persons, insulin can improve memory or cognition when administered intranasally. GLP-1 or its analogues can alter cerebral glucose metabolism. Evidence that long-term outcome can be affected in neurodegenerative diseases by treating these conditions with anti-diabetic drugs or that neurodegenerative diseases may be prevented by rigorous treatment of diabetes is lacking. Conversely, impaired glucose metabolism seems to be more frequent in AD and PD.
Type 2 diabetes, glucose intolerance and obesity all carry an increased risk of cognitive impairment, Alzheimer's disease (AD) or Parkinson's disease (PD). Adjusting for vascular risk factors such as stroke, hypertension or heart disease leaves a substantial residual increased risk for neurodegenerative diseases in diabetes . Relative risk estimates vary considerably, but many report a doubling of the risk for either condition with type 2 diabetes. Type 2 diabetes carries an even higher risks for vascular dementia.
From activated microglia via neuronal damage to brain disease (Click to enlarge)Cognitive decline and overt AD share a number of pathological characteristics with Type 2 diabetes. First, and perhaps most intriguing, insulin signalling is impaired in both conditions; this has led researchers to refer to AD as Type 3 diabetes. Amyloid-beta (Aβ), a culprit protein causing the formation of senile plaques in AD which can also be found in the pancreas in Type 2 diabetes, shares a number of metabolic features with insulin, including binding to the insulin receptor and degradation by insulin degrading enzyme. Further, downstream insulin signals inhibit hyperphosphorylation of intracellular transport (tau) proteins, another hallmark in AD. In PD as well, there seems to be a loss of insulin signalling in the substantia nigra. Low-grade inflammation, oxidative stress and mitochondrial dysfunction are increased in both Type 2 diabetes and neurodegeneration as well as in other chronic diseases .
Although weak, there is an association linking high HbA1c and high glucose variability to declines in cognitive function .
A role for diabetes drugs in neurodegenerative disease?
Although drugs for neurodegenerative diseases are available, most of them alleviating symptoms by manipulating neurotransmitter levels directly, AD and PD remain diseases in search of new efficacious treatments. The epidemiological and molecular relations described above have sparked optimism that drugs or other treatments for metabolic diseases could also affect neurodegeneration.
Life style intervention: Some evidence (animal experiments and epidemiology) exists that calorie restriction and /or exercise may reduce the risk of developing cognitive dysfunction. For overt neurodegenerative diseases there is little evidence for a treatment effect. Bariatric surgery, at the extreme, massively changes the excretion of neuropeptides and gut hormones; several of them have been suggested to be neuroprotective.
Insulin: Pancreatic insulin crosses the blood-brain barrier by a receptor-mediated process. There is a substantial intracerebral production as well. Insulin receptors are abundant in cerebral cortex, hippocampus, hypothalamus, olfactory bulb and cerebellum. Insulin and Aβ compete for the insulin receptor and the degrading enzyme and there is evidence that increasing intracerebral insulin levels can reverse some of Aβ’s effects on neuronal performance. Insulin given as an acute infusion can improve cognition and induce EEG-changes, in healthy individuals and in AD. Intriguingly, insulin given intranasally results in significant increases of insulin levels in the cerebrospinal fluid without causing systemic effects. In pilot experiments intranasal insulin improves memory and cognition in healthy as well as in AD. A large-scale trial in mild cognitive impairment (MCI) and AD, the SNIFF trial, is expected to report outcomes in late 2014 .
Glucagon-like peptide 1: By the incretin effect the gut hormone GLP-1 increases peripheral insulin levels both by binding to β-cell receptors and by stimulating vagal afferents. This, however, is the case only with hyperglycaemia. This and the fact that stable GLP-1 analogues seem to cross the blood-brain barrier and the presence of GLP-1 receptors in cerebral cortex, hippocampus and other regions of the brain has led to a number of preclinical and clinical experiments in both AD and PD models.
In experimental models of either disease GLP-1 is neuroprotective, modulating both Aβ and tau-pathologies in AD models and dopamine secretion in PD models. A common denominator for the effects of GLP-1 has been suggested to be an anti-inflammatory action, reducing the number of activated microglia, which appears an essential feature of all neurodegenerative diseases. An increasing number of observations propose beneficial effects on cognition and learning by GLP-1 in animal models of AD (9). In the clinical setting, GLP-1 can modulate cerebral glucose metabolism in healthy subjects (10), whereas clinical studies in AD or PD are still due to report their first results (Sept 2014).
Metformin: In cell experiments, metformin can decrease tau-formation and in rodents cognition is improved by metformin. There is evidence that metformin crosses the blood-brain barrier and that AMPK-activation occurs also in neurones. Clinical data are lacking (7,8).
Thiazolidinediones: Preclinical studies in AD models suggest that TZDs may be neuroprotective, particularly by reducing inflammation. Further, animal studies suggest improved cognition with these compounds. There has been some debate whether TZDs cross the human blood-brain barrier. Never the less, adequately powered clinical studies with rosiglitazone in AD have been performed, one with rosiglitazone monotherapy, the other with rosiglitazone as add-on to cholinesterase inhibitors. Both studies were negative (7,8).
Leptin and amylin: Increased levels of circulating leptin have been observed in AD. Preclinical experiments suggest that leptin may play roles in the CNS other than those related to energy homeostasis. Thus Aβ production can be reduced by leptin. Likewise, amylin, which crosses the blood brain barrier, possesses a number of neurotrophic effects and appears to be anti-inflammatory. There are no clinical data to support a treatment effect (7,8).
Lessons learnt form insulin resistance and Type 2 diabetes may apply also in the brain, both in neurodegenerative diseases and in depression. A large bulk of evidence is building to connect the CNS and diabetes. Whether the connection will prove useful and provide relief from neurodegenerative diseases or not is currently tested in clinical trials.
^ Shared dysregulated pathways lead to Parkinson's disease and diabetes. Santiago JA, Potashkin JA.
^ Risk of dementia in diabetes mellitus: a systematic review. Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P. Lancet Neurol. 2006 Jan;5(1):64-74.
^ Glucose regulation, cognition, and brain MRI in type 2 diabetes: a systematic review.Geijselaers SL, Sep SJ, Stehouwer CD, Biessels GJ.
^ Repurposing diabetes drugs for brain insulin resistance in Alzheimer disease. Yarchoan M, Arnold SE. Diabetes. 2014 Jul;63(7):2253-61. doi: 10.2337/db14-0287.
^ Type 2 diabetes and cognitive compromise: potential roles of diabetes-related therapies.Kravitz E, Schmeidler J, Schnaider Beeri M.