ntil recently, it was the view that insulin had little effect on
the energy metabolism of the brain. It was thought illogical
that something as important as the brain would be
dependent in any way on when a person last ate a meal. This was
because the brain is a particularly energy intensive organ,
consuming approximately 25% of the body’s total glucose.
However, it was first discovered that astrocytes and microglial cells,
which are major communicating cells in the brain, are insulin sensitive
and later that the insulin receptor had a wide distribution in the brain.
Fluorodeoxy-glucose positron emission tomography studies
demonstrated that insulin had a role in regulating global brain
glucose utilisation in humans, most markedly in the cortical regions.
Other studies have demonstrated that insulin plays a key role in
neuroplasticity, neuromodulation and neurotrophism – the process of
neuronal growth, stimulated by neuronal differentiation and survival.
Epidemiology studies demonstrate that the risk of developing
Alzheimer’s disease is increased by 50% in subjects with diabetes.
Type 2 diabetes, which accounts for 90% of all diabetes, is due to
both insulin resistance and decreased pancreatic
In the early stages, insulin resistance results in the pancreatic
islets secreting more insulin in an attempt to overcome the
resistance. Over time, this leads to an increasing failure of the islet
to produce enough insulin and ultimately hypoinsulinaemia so that
the patient requires insulin injections.
A key question is whether the brain develops insulin resistance.
Recent studies using
brains from patients with
Alzheimer’s, but without diabetes, showed markedly reduced
responses to insulin in the insulin receptor
phosphoinositol-3 kinase signalling pathway in the
hippocampus and to a lesser degree in the cerebellar cortex.
In the hippocampus, the biomarkers of insulin resistance
increased progressively from normal cases through mild cognitive
impairment cases to Alzheimer’s disease regardless of diabetes.
However, there is no evidence that the brain in Alzheimer’s disease
is hyperglycaemic, unlike peripheral tissues in diabetes. Thus an
appropriate term to describe the state of Alzheimer’s disease brain
is ‘insulin-resistant brain’. This is analogous to insulin resistance
syndrome, which is a feature of several peripheral tissue disorders.
Insulin resistance in peripheral tissues could promote insulin
resistance in the brain by reducing brain insulin uptake and by
raising brain levels of A
A second reason why peripheral resistance to insulin may affect
the brain has now been proposed. This is that toxic ceramides
generated by the disturbed lipid metabolism in insulin-resistant
liver pass into the circulation and transit across the blood brain
barrier into the brain. There they induce inflammation, leading to a
second-pronged attack on central insulin action.
These studies lead to the view that drug treatments that improve
either or both central and peripheral insulin resistance are
potential treatments, or at least agents, to delay the progression
of mild cognitive impairment to Alzheimer’s disease. In fact, two
anti-diabetes drugs have already been shown to have some
beneficial effects. They are metformin and thiazolidinedione
rosiglitazone, and a third drug, pioglitazone, has been proposed
for a large clinical trial.
However, none of these drugs are a perfect treatment. Many
patients do not easily tolerate metformin as a result of
gastrointestinal side effects and the thiazolidinedione insulin
sensitiser drugs produce weight gain and water retention. There is,
therefore, a need for a concerted effort to discover drugs that act
to cause insulin sensitisation in the brain and liver. Such drugs are
a potential treatment for Alzheimer’s patients possibly when
coupled with nasal delivery of insulin, which has been shown to
improve learning and memory in clinical trials. The greater need,
however, is to develop strategies that will delay the onset and
progression of mild cognitive impairment towards Alzheimer’s.
For this, drugs are not the answer, unless one had diagnostic
assays that were highly predictive, since it would be unethical to
give drugs with inherent risks to people who might not develop the
disease. However, if non-toxic plant extracts with similar insulin
sensitiser action could be identified, these could be made
available to the public.
University of Buckingham
te l :
+44 (0)1280 email@example.com www.buckingham.ac.uk/clore/
Several lines of evidence suggest that Alzheimer’s disease has strong links with
insulin action and that the condition could be described as ‘diabetes of the brain’
Alzheimer’s and diabetes
The Buckingham Institute of Translational Medicine
The University of Buckingham