|
As
life spans lengthen and the elderly population rises,
diabetes and Alzheimer’s disease (AD) have come to
afflict rapidly increasing numbers of people. The two
diseases have entirely different symptoms and
etiology, but now new evidence suggests they are
causally linked. If that link proves true, some drugs
might find use in not just one, but two of the
world’s largest pharmaceutical market arenas.
Epidemiological
evidence shows that signaling pathways involving
insulin — which regulates blood glucose — also
generate the brain lesions associated with AD. At the
International Conference on Alzheimer’s Disease (ICAD)
in Madrid in July, Swedish researchers published the
results of a nine-year study of 1,173 elderly people.
They found that individuals with borderline diabetes
were at a 70 percent higher risk of developing
dementia and AD than nondiabetics.
The
physiological reasons for this are gradually being
unraveled. Examined post-mortem, the brains of
Alzheimer’s patients show two pathological
hallmarks: extracellular “beta-amyloid plaques”
and intracellular “neurofibrillary tangles.” Both
these abnormal protein structures seem to be deposited
in the brain’s cortex and hippocampus just as the
neurodegeneration associated with AD dementia appears
— although no one really knows what is cause and
what is effect, says Calum Sutherland, a senior
neuroendocrinologist at Scotland’s Dundee
University.
However,
what is known is that formation of these plaques and
tangles is promoted by the enzyme glycogen synthase
kinase 3 (GSK-3, well known to be vital in
carbohydrate metabolism). GSK-3 acts by
phosphorylating two key brain proteins called tau and
CRMP-2 (collapsin response mediator protein 2),
causing them to precipitate into clumps.
PLAQUE
ATTACK: (Left) Insulin signaling in healthy
individual with normal beta-amyloid metabolism; and
(Right) in individual with insulin resistance, leading
to plaque/tangle formation in brain. (Source: Calum
Sutherland) Click
here to enlarge
This
mechanism is closely related to diabetes because
insulin — present in high levels in the brains of
diabetics — is a powerful inhibitor of GSK-3 action.
“A very important part of insulin’s action in
controlling blood glucose is that, when it binds to a
cell receptor, it turns off GSK-3 action inside the
cell,” says Sutherland. The implication, he says, is
that compounds that mimic insulin by reducing GSK-3
activity could be therapeutic in AD.
Such
GSK-3 inhibitors have been developed, albeit initially
for diabetes treatment, notes Sutherland. But their
pharmacological drawback is that GSK-3 has many
substrates, so blocking it could produce a range of
unwanted side effects. “In diabetes we have to treat
people continuously for many decades, so GSK-3
inhibitors have tended to get stuck in toxicology,”
he says. But toxicology isn’t such an issue when
treating AD, which usually occurs late in life and for
which very few alternative therapies exist.
The
question is whether inhibiting GSK-3 — and thus
reducing plaque and tangle formation — will actually
prevent or reverse the effects of Alzheimer’s in
humans. Signs from preclinical are good. In May,
Spanish researchers reported reversing
neurodegeneration in transgenic mice by
down-regulating the mice brain GSK-3 levels. The mice
they used were genetically engineered to overexpress
GSK-3 in the forebrain, but this overexpression could
be “switched off” by giving the mice antibiotics.
In the experiment, with GSK-3 overexpression switched
on, the mice at first developed a form of dementia
similar to AD, soon becoming unable to recognize
familiar objects. But when overexpression was switched
off and their brain GSK-3 levels fell back to normal,
their memories recovered.
The
authors say this proves that “tau
hyperphosphorylation, apoptotic neuronal death, and
cognitive deficit attributable to increased
hippocampal GSK-3 activity can be completely reverted
by the restoration of normal GSK-3 activity by
silencing of transgene expression” (Engel, T. et al.
J Neurosci 26, 5083-90; 2006). “It strongly
supports the notion that GSK-3 inhibitors are a good
therapeutic target for AD.”
Clinical
Candidates
The
only GSK-3 inhibitor known to have entered the clinic
for AD has been developed by Neuropharma, a subsidiary
of Spain's leading drug company, Zeltia.
Neuropharma’s lead compound, called NP031112, is a
heterocyclic molecule of the thiadiazolidinone (TDZD)
class. It has already been given to 30-plus subjects
in a German Phase I safety assessment and dose
escalation study, the company announced at ICAD. No
serious adverse events have been noted, and further
studies are planned.
But
if GSK-3 is a promising target, why not attack insulin
signaling itself? Sutherland says that already-known
insulin sensitizers, such as metformin or the
thiazolidinediones, could well have therapeutic
effects on AD.
Pharma
companies with insulin sensitizer drugs on the market
for diabetes are already working on this idea. Earlier
this year, GlaxoSmithKline published the results of a
Phase II study of rosiglitazone (Avandia) in about 100
AD patients. It was half successful, says Mark
Strachan, diabetes and endocrinology consultant
at Western General Hospital, Edinburgh ,
U.K. “One genetic subgroup — those without the
apolipoprotein-E4 (ApoE4) allele — did better with
Avandia,” he explains. “But those with ApoE4 did
slightly worse.”
This
result was not unexpected: epidemiological
data, including the Swedish study reported at ICAD,
have shown that ApoE4 somehow combines with insulin to
cause AD brain lesions. And one 2002 post-mortem study
of several hundred Alzheimer’s brains found that
brain tangles and plaques are more common in people
with diabetes, but only if they are also ApoE4
positive.
Armed
with its clinical data, GlaxoSmithKline is now
preparing some major multi-center trials of Avandia in
AD, which the company’s CEO, Jean-Pierre Garnier,
recently described as “absolutely huge, if it works
out.” But GlaxoSmith-Kline is not alone. Also at
ICAD was reported the results of a small-scale
U.S. study of pioglitazone (Actos, Takeda) for
nondiabetics with AD. The treatment appeared to reduce
Alzheimer’s progression, say the researchers at
University of Virginia and Case Western, but the
numbers were too small to be conclusive — perhaps
because the patients were not stratified by ApoE
genotype. Lead researcher David Geldmacher says the
results are promising enough to justify a larger
study.
A
TANGLED WEB: Insulin resistance appears to
trigger critical steps in neurodegeneration. (Source:
Calum Sutherland) Click
here to enlarge
The
ApoE4 Effect
Why
does genotype matter? Strachan notes the mechanism
relating ApoE to Alzheimer’s is not understood. But
one clue could be in a recent histopathological study
that found much lower expression of insulin-degrading
enzyme (IDE) in the brains of AD patients with the
ApoE4 allele.
The
theory is that, in the normal brain, IDE breaks down
beta-amyloid peptide before it accumulates, thus
delaying the formation of senile plaques and
protecting against AD. In diabetics, however, the
large amounts of insulin present in the brain compete
with IDE for binding sites on the amyloid, thus
removing the protection given by IDE, and allowing
amyloid plaques to form.
This
suggests that IDE is yet another possible target for
Alzheimer’s therapies, although the hypothesis still
has to be confirmed clinically, says Geert Biessels of
Utrecht University in the Netherlands: “There are
still many loose ends; we know relatively little on
how diabetes and its treatment affect cerebral insulin
and its receptor.”
Sutherland
is doubtful. “It’s difficult to see how you could
target IDE, except by trying to overexpress it by gene
therapy or by making a form that specifically targets
beta-amyloid better than insulin,” he says. “If
instead we had a general body insulin sensitizer, we
could lower brain insulin levels so the IDE would
automatically be available to degrade beta-amyloid
peptide. That’s the Holy Grail — but it will
come.”
Alzheimer’s
affects more than 27 million people worldwide, and
that number is soaring. Current therapies such as
cholinesterase inhibitors only treat the symptoms and
cannot prevent continued degeneration. So if drug
discovery research originally aimed at diabetes does
yield therapies for AD, it would be a major
breakthrough. And even if it turns out that glitazones
don’t act directly on plaque formation, they could
still help protect brain function in diabetics simply
by controlling blood sugar and thus avoiding the nerve
damage caused by hyperglycemia, Strachan points out.
“Clearly,
if the results are positive, they have major potential
for GlaxoSmith-Kline,” says Strachan. “And if
Takeda is interested as well, you can bet that several
other companies are looking at it too.”
Email
this page to a friend
|
