People who have diabetes are at elevated risk for Alzheimer’s (see AlzRisk entry). Why is that? In the April 13 Journal of Neuroscience, researchers led by Paul Mathews, Nathan S. Kline Institute for Psychiatric Research, Orangeburg, New York, report that diabetes causes insulin resistance in the brains of non-human primates, which then activates kinases that phosphorylate tau and downregulate proteins that degrade Aβ. The result is widespread increase in phosphorylated tau in the brain and more soluble Aβ in the hippocampus. “These are consistent with very early events that would predispose the brain to developing Alzheimer’s over time,” said Mathews. “That idea definitely warrants additional study.”

Rodent studies have hinted that altered insulin signaling in diabetes causes a rise in tau phosphorylation, though researchers don’t know why (for review, see El Khoury et al., 2014). What happens to Aβ is even less clear, Mathews said. Could monkeys offer clues?

Five years ago, scientists examined the peripheral effects of type 1 diabetes on 25 middle-aged male vervet monkeys, aged six to eight years (see Kavanaugh et al., 2011; Saisho et al., 2011). The researchers had used the toxin streptozotocin (STZ) to kill off insulin-producing β-cells of the pancreas in 15 of the animals, while the other 10 served as controls. For up to five months, two daily insulin injections kept blood glucose relatively steady in the STZ-treated animals, modeling controlled diabetes in people. Then the monkeys were killed and their tissue stored. None were tested for cognition or appeared to be demented at that time.

When Stephen Ginsberg, co-author of the current study, found out that the brain tissues were preserved, he jumped at the chance to examine them. While monkeys can model human disease more closely than do transgenic mice that overproduce Aβ, they are costly and research with them is tightly regulated by ethics rules. “We were lucky that we could make use of tissue that existed from a prior, well-conducted study,” said Mathews.

First author Jose Morales-Corraliza prepared homogenates from frozen samples of hippocampi, frontal cortices, superior temporal cortices, and cerebella. He first looked for signs that insulin dysregulation outside the brain had caused insulin resistance inside it. Sure enough, the brain insulin receptor substrate 1 (IRS1) had been phosphorylated and inactivated in all four brain regions of the STZ-treated animals, suggesting that their brain cells were more resistant to insulin. A similar bump in inhibitory IRS1 phosphorylation has been reported in people progressing from mild cognitive impairment to AD (Talbot et al., 2012). 

Did that resistance affect tau? The results suggest yes. Total tau matched control levels, but western blots with antibodies PHF-1 and CP13 detected more phosphorylated tau in all four sections of the brain. The group measured the phosphorylation status of kinases that typically phosphorylate the protein as well, and found more of the activated form of ERK1/2.

Morales-Corraliza and colleagues used ELISAs to look for changes in soluble Aβ that might drive regional patterns of pathology. In the hippocampus, they found twice as much Aβ in STZ-treated animals than in controls, and a 40 percent rise in the temporal cortex. Neither the levels of APP or its β-secretase fragments were higher, suggesting APP processing was normal. Levels of the RAGE and LRP1 proteins that ferry Aβ across the blood-brain barrier were also unaltered by the chronically induced diabetic state. However, expression of the Aβ-degrading enzyme neprilysin was lower in the hippocampus of the diabetic monkeys. “Over decades, this could drive Aβ accumulation and pathology,” said Mathews. A localized rise in Aβ could also drive tau pathology seen in temporal lobe structures, he proposed.

Overall, this study points to early changes that might predispose diabetic people to AD, said Mathews. The authors acknowledged that type 2 diabetes is a more common risk factor for AD than type 1. However, type 1 diabetes has also been linked with AD and in both diseases disruption of insulin signaling leads to insulin resistance.

This study gives another clue to how AD and diabetes are linked, said Emmanuel Planel, Centre Hospitalier de l'Université Laval, Quebec. It would be a leap to assume the same changes in Aβ and tau happen in humans, but researchers could next look for loss of neprilysin and activated ERK1/2 in people with type 1 and type 2 diabetes, Planel said. He found it remarkable that these changes occurred despite treatment with insulin. Planel praised the comparatively large number of monkeys used in this study.

Cora O’Neill, University College, Cork, Ireland, said that the study clearly suggests peripheral insulin resistance can cause changes to Aβ and tau in the brain. However, she pointed out that people with AD can develop insulin resistance in the brain regardless of whether they also have diabetes. Understanding how insulin resistance develops in AD may enable scientists to target those pathways, normalize them, and with that perhaps ameliorate cognitive dysfunction, O’Neill said. She noted that it would have been nice to study these monkey brains at the cellular level to see if the same cells that were resistant to insulin also harbored the phosphorylated tau, and to better understand the relationship between Aβ and tau in these brains.

The study fits with the concept of insulin signaling failure as an early trigger for brain damage, including Alzheimer’s, said Christian Holscher, Lancaster University, U.K. Even if these changes don’t by themselves lead to Alzheimer’s down the road, they may weaken the neurons and make the brain more vulnerable to other stressors that then lead to disease. “This could explain why diabetes is a risk factor for AD,” he said.—Gwyneth Dickey Zakaib


  1. I really like this paper because it demonstrates, in a primate model, that STZ-diabetes causes molecular/biochemical abnormalities that are characteristic of AD and strengthens the hypothesis that AD is a variant of diabetes mellitus (DM). This work represents a continuation of the story in which similar findings were reported in rat and I believe also mouse models. A few things missing from the work raise questions. What was the neuropathology in these monkeys? Did they have any cognitive or behavioral abnormalities? Did they have any vascular amyloid? And did authors exclude the possibility that STZ crossed the blood-brain barrier and directly caused CNS disease? The point about intracerebroventricular STZ is to show that even without type 1 (T1DM) or type 2 diabetes (T2DM), exactly the same abnormalities seen in AD can be produced. The bigger question raised by our group some time ago is, what is the link between STZ and AD? High fat/caloric diets produce T2DM, but brain abnormalities, although similar to AD, are not as severe. In addition, individuals with T1DM get cognitive impairment, but not with the severity seen in STZ models. In other words, I believe this paper helps greatly in supporting the overall concept but also indicates the need to dig deeper into the specific mechanisms of disease. For example, the authors didn't address the role of inflammation, which is part of DM and an early abnormality in AD.

Make a Comment

To make a comment you must login or register.


Antibody Citations

  1. Tau phos Ser396/Ser404 (PHF-1)
  2. Tau phos Ser202 (CP13)

Paper Citations

  1. . Insulin dysfunction and Tau pathology. Front Cell Neurosci. 2014;8:22. Epub 2014 Feb 11 PubMed.
  2. . Characterization and validation of a streptozotocin-induced diabetes model in the vervet monkey. J Pharmacol Toxicol Methods. 2011 May-Jun;63(3):296-303. Epub 2011 Feb 26 PubMed.
  3. . Ongoing beta-cell turnover in adult nonhuman primates is not adaptively increased in streptozotocin-induced diabetes. Diabetes. 2011 Mar;60(3):848-56. Epub 2011 Jan 26 PubMed.
  4. . Demonstrated brain insulin resistance in Alzheimer's disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. J Clin Invest. 2012 Apr;122(4):1316-38. PubMed.

External Citations

  1. AlzRisk

Further Reading


  1. . Diabetes mellitus accelerates Aβ pathology in brain accompanied by enhanced GAβ generation in nonhuman primates. PLoS One. 2015;10(2):e0117362. Epub 2015 Feb 12 PubMed.
  2. . Brain Insulin Signaling Is Increased in Insulin-Resistant States and Decreases in FOXOs and PGC-1α and Increases in Aβ1-40/42 and Phospho-Tau May Abet Alzheimer Development. Diabetes. 2016 Jul;65(7):1892-903. Epub 2016 Feb 19 PubMed.
  3. . How does diabetes accelerate Alzheimer disease pathology?. Nat Rev Neurol. 2010 Oct;6(10):551-9. PubMed.

Primary Papers

  1. . Brain-Wide Insulin Resistance, Tau Phosphorylation Changes, and Hippocampal Neprilysin and Amyloid-β Alterations in a Monkey Model of Type 1 Diabetes. J Neurosci. 2016 Apr 13;36(15):4248-58. PubMed.