Two new papers add to growing evidence linking diabetes and neurodegenerative disease, including dementia and Parkinsonism. In the January 12 Archives of Neurology, researchers led by Suzanne Craft at the VA Puget Sound and University of Washington, Seattle, report that dementia patients with and without type 2 diabetes have subtle differences in their dementia pathology and provide a hint that treating diabetes might lessen the dementia pathology in people with both diseases. Supporting this general notion, a report in the January 21 PNAS online shows that a therapy approved for type 2 diabetes prevents neurodegeneration in some cellular and animal models of stroke and Parkinson disease. Researchers led by Nigel Greig at the National Institute on Aging in Baltimore, Maryland, report that exendin-4, an agonist of the glucagon-like peptide 1 (GLP-1) receptor, limits stroke damage and prevents chemical-induced Parkinson pathology in models. Together, the papers suggest that not only is diabetes a risk factor for neurodegenerative disease, but that treating the metabolic disorder could pay added dividends in the central nervous system.

Whether GLP-1R agonists can be finessed to fit that bill is not clear. Exendin-4 (Ex-4) is the first of a new class of GLP-1 receptor agonists, called incretin mimetics, that has been approved for the treatment of type 2 diabetes. However, the agonist and GLP-1 itself are peptides that do not readily cross the blood-brain barrier, according to Matthew During at Weill Medical College of Cornell University, New York. During was not involved in the present study, but has shown previously that GLP-1 is neuroprotective and improves learning and memory in mice (see ARF related news story). “I think if there was a [CNS] active form that could be administered systemically, a small molecule or truncated peptide, then this class of drugs could be valuable for a range of neurodegenerative disease, including Parkinson’s and Alzheimer’s, because GLP-1 clearly supports and boosts the ability of these neurons to deal with insults,” During told ARF.

In the new study, Greig and colleagues report that GLP-1R agonists protect neurons in both cell and animal models of stroke and Parkinson disease. The strategy could prove useful in AD, too, since the authors previously showed that Ex-4 protects hippocampal neurons from Aβ toxicity and also reduced brain Aβ levels in a cholinergic forebrain lesion model (see Perry et al., 2003).

Ex-4 has pleiotropic effects in the pancreas, where it induces release of insulin and protects β cells from apoptosis (see Li et al., 2003). Greig and colleagues were the first to show that GLP-1 and Ex-4 also have neurotrophic effects in the brain (see Perry et al., 2002). Now, first author Lazhou Li and colleagues show that the GLP-1 receptor (GLP-1R) is present in primary cortical and ventral mesencephalic neurons, and functionally activated by GLP-1. They further report that both GLP-1 and the longer-acting Ex-4 protect cultured cortical neurons against hypoxia and dopaminergic neurons against 6-hydroxydopa, a chemical that can induce Parkinson disease (PD) in mammals. In vivo, Ex-4 pre-treatment reduced the size of the infarct in an artery occlusion model of stroke and protected against dopaminergic neuron loss when given prior to MPTP, a mitochondrial toxin that causes PD.

“Here we focused on stroke and Parkinson disease because they epitomize acute and chronic neurodegenerative diseases,” said Greig, adding that the strategy could work in other diseases, too. “Studies by Suzanne Craft and others show that there seems to be, if not frank diabetes, at least a large amount of glucose intolerance associated with all the neurodegenerative diseases.” He noted that one advantage of using GLP-1R agonists is that their action is dependent on glucose concentration; in other words, the agent induces insulin release only when glucose levels are high. “That’s nice because it can be worrying to treat cognitively impaired patients with glucose-lowering drugs,” said Greig.

During cautioned that systemic consequences would have to be evaluated carefully. “The hope is to find one [GLP-1R agonist] that is systemically active and that doesn’t have potent effects on peripheral metabolism. You will always have to fight that, even though it is true that GLP-1R agonists have a more profound effect if you are diabetic,” he said.

It is not clear if anyone is planning to study Ex-4 in clinical trials of neurodegenerative disease. Apart from their safety, their effectiveness in people who already have these conditions would need to be established. In the published experiments, the agonist was always given prior to the pathological insult, i.e., in a preventive mode. “There is an increased incidence of stroke and a suggested increase in Parkinson’s in people with diabetes. From that perspective, in several years’ time it might be interesting to look back and see if the incidence of stroke and a variety of neurodegenerative conditions is lower in patients who have been on Byetta® for a period of time,” suggested Greig (Byetta® is the brand name for Ex-4).

As for how Ex-4 protects neurons, the mechanism seems to be tied to the G proteins coupled to the GLP-1R receptor and their induction of cyclic AMP, which is neuroprotective. “This is a new therapeutic strategy to address neurodegenerative conditions. It is not going after tau, α-synuclein, or even Aβ. It is really going after cell survival,” said Greig.

The general finding that patients with diabetes are at greater risk for dementia (see, for example, data from a recent study on twins Xu et al., 2009) draws further support by new data from Craft and colleagues. These researchers report that the underlying pathology in people with dementia is different between those who have or do not have diabetes mellitus, as well. This reporter first noted these findings when they were presented at the Society for Neuroscience meeting in San Diego in 2007 (see ARF related news story). First author Joshua Sonnen and colleagues carried out detailed pathological autopsies on people who enrolled in a community-based study of incident dementia called the Adult Changes in Thought Study and compared dementia-related pathology in patients who had died with (n = 59) or without (n = 137) a diagnosis of diabetes mellitus.

The scientists found that dementia patients had the same degree of neurofibrillary tangles regardless of their diabetes status. But dementia patients without diabetes had a significantly higher amyloid plaque load as judged by thioflavin S staining. These same patients also had significantly higher levels of formic-acid extractable Aβ in the brain compared to dementia patients with diabetes. And the dementia/diabetes-free group had the highest degree of cerebral amyloid angiopathy (CAA). Next in line in terms of CAA severity were the dementia-and-diabetes patients; followed last by non-demented controls with or without diabetes. “Together, these data support the conclusion that patients with diabetes mellitus and dementia die with a lower Aβ burden in brain parenchyma and cerebral blood vessels compared with their counterparts who die with dementia but do not have diabetes,” write the authors. The data suggest to the authors that having diabetes mellitus makes patients more vulnerable to dementia (see Q&A below with Craft).

It is not amyloid pathology that puts people with diabetes at higher risk of dementia. What, then, could it be? A new clue comes from studying the brain vasculature and the two biomarkers interleukin 6 (IL-6) and F2-isoprostanes, which accompany inflammation and oxidative damage, respectively. The Seattle researchers found that patients with both diseases had the greatest numbers of deep cerebral microvascular infarcts and higher IL-6 levels, both pointing to inflammation. Dementia patients without diabetes, on the other hand, had higher levels of F2-isoprostanes in the brain, suggesting a higher degree of free radical damage. “These novel characterizations of two apparently different patterns of injury in dementia depending on diabetes mellitus status may have important etiologic and therapeutic implications,” conclude the authors.

On treatment, the researchers noted that dementia-and-diabetes patients who were on diabetic medication had a lower plaque load than such patients whose diabetes was untreated (see also Beeri et al., 2008). That would seem to suggest that treating diabetes may help reduce some AD pathology. Interpretation of these data is not trivial, however, since dementia-free diabetes patients have higher plaque loads if they are being treated for diabetes. Reconciling these contradictory effects might come down to the severity of the diabetes, the authors suggest.—Tom Fagan.

Sonnen JA, Larson EB, Brickell K, Crane PK, Woltjer R, Montine TJ, Craft S. Different patterns of cerebral injury in dementia with or without diabetes. Arch Neurol. 2009, January 12 online publication. Abstract

Li Y, Perry T, Kindy MS, Harvey BK, Tweedie D, Holloway HW, Powers K, Shen H, Egan JM, Sambamurti K, Brossi A, Lahiri DK, Mattson MP, Hoffer BJ, Wang Y, Greig NH. GLP-1 receptor stimulation preserves primary cortical and dopaminergic neurons in cellular and rodent models of stroke and Parkinsonism. PNAS 2009, January 21. Abstract

Q&A with Suzanne Craft. Questions by Tom Fagan.

Q: One of the main findings from this study is that on autopsy, dementia patients with diabetes mellitus (DM) have lower plaque burden than dementia patients without DM. One interpretation from this is that DM patients are more susceptible to dementia because they are demented at lower plaque loads. Could you elaborate on this?
A: DM patients may have other pathological processes that contribute to dementia expression. Vascular injury is an obvious candidate supported by this and other studies. Vascular injury and amyloid load may have additive effects, and thus cognitive and functional impairment become apparent at lower levels of plaque burden.

Q: Keeping to the same theme, could it be that dementia in DM patients is just not as advanced? Is it possible that the patients may be dying of DM-related co-morbidities earlier than they would if they had no DM, and that’s why the plaque burden is lower?
A: In our study, the level of dementia as indicated by CASI score was similar for the diabetic and non-diabetic dementia cases, as was age at death. So it does not appear that the reduced plaque burden reflected less advanced dementia or differential morbidity.

Q: As you mentioned in your paper, some previous studies have not seen any correlation between AD pathology and DM. How do your data fit with those observations?
A: With regard to Aβ, our study is largely consistent with those findings. Our study takes those findings one step further, however, to suggest that the absence of AD pathology may be associated with treatment for DM. The DM cases with dementia who did not receive treatment had levels of plaque burden similar to non-DM dementia cases. We must interpret these data cautiously, because untreated DM cases with dementia had less severe DM than did treated cases, although one might have predicted that less severe DM would be associated with less, not greater, plaque burden if the association was due to factors other than treatment.

A second point worth noting is that neurofibrillary tangles as indexed by Braak staging were comparable among all dementia groups, regardless of DM or DM treatment status, so high NFT burden was an invariant feature of dementia in our sample.

Q: Do you think dementia + DM patients might perform differently from dementia patients with no DM if you put them through some sophisticated cognitive or neuropsychological tests? Might such testing reveal differences in pathology? Could different areas of the brain be affected in DM+ patients, for example?
A: Yes, based on cognitive studies in DM, as well as on typical vascular injury patterns (e.g., ischemic patterns), one might predict they would have frontal impairment, which may not be adequately assessed in instruments like the CASI. We would not be able to assess that possibility in our sample.

Q: Is there something about having DM that makes patients more vulnerable to the pathology associated with plaques or tangles? If so, do you have any thoughts on how that vulnerability might be realized, mechanistically?
A: There are a number of pathologies associated with DM that might interact with amyloid and tau pathology. Peripheral insulin resistance and hyperinsulinemia are associated with reduced transport of insulin into the CNS, which may deprive the brain of the salutory actions of insulin on synaptic maintenance and remodeling, neuronal repair, and other trophic processes needed to resist the neurotoxic effects of amyloid. DM is associated with increased advanced glycation end-products, inflammation, and macro- and microvascular injury, each of which can make the brain more vulnerable to amyloid or tau pathology.

Q: Accepting that DM patients are more susceptible to dementia, do you think that is tied in with cerebrovascular damage?
A: Yes, I believe that vascular injury is often contributory, although the precise mechanisms through which vascular dysfunction increases dementia susceptibility in DM remain in question, and are an exciting focus of current research. Careful postmortem studies indicate that isolated vascular pathology is the least frequent neuropathological pattern observed in late-life dementia, and for most patients, markers of vascular injury coexist with traditional AD hallmarks. In some cases, the AD hallmarks may conceivably be promoted by a specific form of vascular injury; for example, BBB dysfunction may affect Aβ transport between brain and periphery, and thereby contribute to parenchymal and neurovascular amyloid deposition. Conversely, exposure to chronic Aβ elevations may conceivably compromise vascular function, resulting in ischemia or microvascular injury.

DM is caused in most cases by insulin resistance, which has many negative effects on vascular function. These effects may be directly related to impaired insulin action, as well as caused by insulin resistance-induced dyslipidemia and inflammation. Insulin directly affects vasoreactivity and hemodynamic functions, such as capillary recruitment, vasodilation, and regional blood flow. Thus, insulin resistance negatively impacts vascular function via a number of mechanisms, and vascular dysfunction likely plays an important role in dementia associated with DM.

Q: Inflammation seems to be a major consideration in the DM patients, as seen by increased IL-6 levels and microvascular damage. Does this suggest any particular therapeutic approach?
A: I would predict that approaches that address the underlying insulin resistance would in turn reduce inflammation. These strategies would include lifestyle modifications to improve diet, increase physical activity, and reduce adiposity, as well as pharmacologic agents such as PPAR agonists and other insulin sensitizers. The sirtuins are another interesting class of compounds with insulin sensitizing and anti-inflammatory effects.

Q: Clinical trials of anti-inflammatories have proven disappointing. Do you think that data need to be re-examined? Where people with DM were included in trials, would you propose reanalyzing those data based on DM status? Are new anti-inflammatory trials in DM patients warranted?
A: Chronic anti-inflammatory administration may have unanticipated adverse effects. Although speculative, one possibility is that such blanket inhibition prevents the “beneficial” inflammatory response needed when the body or brain encounters an immune challenge. For example, metabolism of food after even a healthy meal provokes a mild, temporary inflammatory response, and high-fat meals provoke more extreme inflammation. Approaches aimed at the cause of the chronic inflammation in DM, which is likely due to the underlying insulin resistance, chronic hyperinsulinemia, obesity, and dyslipidemia, may be preferable.


Make a Comment

To make a comment you must login or register.

Comments on this content

No Available Comments

Comments on Primary Papers for this Article

No Available Comments on Primary Papers for this Article


News Citations

  1. Learning and Neuroprotective Role Proffered for Glucagon-Derived Peptide
  2. Diabetes-Insulin Roundup: Dementia Connection Grows Stronger, Part 2

Paper Citations

  1. . Glucagon-like peptide-1 decreases endogenous amyloid-beta peptide (Abeta) levels and protects hippocampal neurons from death induced by Abeta and iron. J Neurosci Res. 2003 Jun 1;72(5):603-12. PubMed.
  2. . Glucagon-like peptide-1 receptor signaling modulates beta cell apoptosis. J Biol Chem. 2003 Jan 3;278(1):471-8. PubMed.
  3. . A novel neurotrophic property of glucagon-like peptide 1: a promoter of nerve growth factor-mediated differentiation in PC12 cells. J Pharmacol Exp Ther. 2002 Mar;300(3):958-66. PubMed.
  4. . Mid- and late-life diabetes in relation to the risk of dementia: a population-based twin study. Diabetes. 2009 Jan 1;58(1):71-7. PubMed.
  5. . Insulin in combination with other diabetes medication is associated with less Alzheimer neuropathology. Neurology. 2008 Sep 2;71(10):750-7. PubMed.
  6. . Different patterns of cerebral injury in dementia with or without diabetes. Arch Neurol. 2009 Mar;66(3):315-22. PubMed.
  7. . GLP-1 receptor stimulation preserves primary cortical and dopaminergic neurons in cellular and rodent models of stroke and Parkinsonism. Proc Natl Acad Sci U S A. 2009 Jan 27;106(4):1285-90. PubMed.

Other Citations

  1. see Q&A below with Craft

External Citations

  1. Adult Changes in Thought Study

Further Reading


  1. . Different patterns of cerebral injury in dementia with or without diabetes. Arch Neurol. 2009 Mar;66(3):315-22. PubMed.
  2. . GLP-1 receptor stimulation preserves primary cortical and dopaminergic neurons in cellular and rodent models of stroke and Parkinsonism. Proc Natl Acad Sci U S A. 2009 Jan 27;106(4):1285-90. PubMed.


  1. Diabetes-Insulin Roundup: Dementia Connection Grows Stronger, Part 2
  2. Learning and Neuroprotective Role Proffered for Glucagon-Derived Peptide
  3. Diabetes-Insulin Roundup: Dementia Connection Grows Stronger, Part 1

Primary Papers

  1. . Different patterns of cerebral injury in dementia with or without diabetes. Arch Neurol. 2009 Mar;66(3):315-22. PubMed.
  2. . GLP-1 receptor stimulation preserves primary cortical and dopaminergic neurons in cellular and rodent models of stroke and Parkinsonism. Proc Natl Acad Sci U S A. 2009 Jan 27;106(4):1285-90. PubMed.