Transcript TARGETING THE CREB PATHWAY FOR MEMORY ENHANCERS

TARGETING THE CREB
PATHWAY
FOR MEMORY ENHANCERS
Tim Tully*, Rusiko Bourtchouladze*,
Rod Scott* and John Tallman*
Helicon Therapeutics and Cold Spring
Harbor Laboratory
• How are potential molecular targets for
drugs found?
• How are potential memory drugs
discovered and evaluated?
• What memory-related diseases are good
candidates for CREB-related therapy?
•
•
•
•
•
•
ACETYLCHOLINE, GLUTAMATE
the neurotransmitters acetylcholine and glutamate
CHOLINERGIC
Pertaining to acetylcholine neurotransmission.
CHOLINESTERASE
An enzyme that metabolizes the neurotransmitter
acetylcholine. Cholinesterase inhibitors lead to increased
levels of acetylcholine in the brain.
• GLUTAMATERGIC
• Pertaining to glutamate neurotransmission.
• Alzheimer’s disease (AD) pathology
– associated with a progressive loss of
cholinergic neurons
– historically the pharmaceutical industry has
focused on the identification of cholinesterase
inhibitors.
• Marketed AD drugs such as donzepil,
rivastigmine and galantamine yield mild
improvements
– effects are variable and not long-lasting
• The biochemical basis of memory.
• NMDA (N-methyl-D-aspartate)-receptor
activation (in mammals) seems to be central to
the local, transient biochemical response at the
synapse.
• These transient biochemical responses induce
changes in gene expression in the cell nucleus
via activation (that is, phosphorylation) of the
transcription factor cAMP-response-elementbinding protein (CREB).
• This transcriptional response also depends on
NMDA receptor activation.
• Protein kinase A (PKA) and mitogen-activated
protein (MAP) kinase signalling pathways play
dominant roles in activation.
• CREB regulates a transcription factor cascade
ultimately involved in a growth process that
yields synapse-specific structural changes.
• This process seems to involve microtubuledependent transport of nascent mRNAs to
synaptic regions and local regulation of
translation.
• A molecular switch for LTM.
• Neurogenetic studies have shown CREB
to be a key control point for LTM formation.
• Loss-of-function manipulations of CREB
leave learning and STM intact, but impair
LTM.
• Gain-of-function manipulations leave
learning and STM intact, but enhance LTM
formation specifically by reducing the
amount of training required to produce
maximal LTM.
• Similar manipulations of CREB also
produce changes in synaptic structure and
function in several animal models and in
various regions of the mammalian brain.
• Molecular targets for drug discovery are those,
such as CREB, which act as molecular switches.
• many genes are involved in stages of memory
formation.
• likely several genes will act as molecular
switches and are potential targets for drug
screening.
• combinations of drugs may provide more
effective treatments.
Two classes of drug targets
• a-amino-3-hydroxy-5-methyl-4-isoxazole
propionicacid (AMPA) receptors
• genes in the CREB pathway
• AMPA receptors depolarize postsynaptic
membranes in response to glutamate
• Modulators of AMPA receptors have been
developed on the basis of the expectation that
they might facilitate NMDA receptor-dependent
induction of LTP and, thereby, the acquisition of
new memories.
• Several of these compounds have yielded
encouraging results from preclinical animal
testing and are now undergoing clinical
evaluations.
CREB / cAMP genes
• ADENYLYL CYCLASE (AC).
– An enzyme that synthesizes cAMP from ATP
and is involved in intracellular signalling,
usually after neurotransmitter activation.
• PHOSPHODIESTERASE (PDE).
– An enzyme that hydrolyzes cAMP into AMP.
• Screen for enhancers of CREB signalling.
• molecular target was discovered to be
PHOSPHODIESTERASE4 (PDE4).
Assay for CREB-based memory
enhancement.
• naive mouse is placed into a novel chamber containing
distinct visual, olfactory and tactile cues.
• mouse receives mild electroshock to its feet.
• mouse will remember for sometime afterwards that the
chamber is dangerous.
• When placed back into the chamber at a later time, the
mouse will freeze, sitting stone still for many seconds,
which is its natural response to danger.
• percentage of time during an observation period that the
mouse spends frozen represents a quantitative measure
of its memory of a dangerous place.
• normal mouse will learn this task in just one
training trial.
• Tully modified procedure so that more training
trials were required to form maximal LTM.
• By reducing the intensity of footshock, four-day
memory retention in normal mice (C57/Bl6) was
near zero after one training trial and required
five training trials to reach maximum levels.
• In initial experiments with drugs, animals were
cannulated and drug (or vehicle alone) was
injected directly into the hippocampus shortly
before training.
• Mice received two (submaximal) or five
(maximal) training trials.
• A number of Helicon’s PDE4 inhibitors, and the
prototypic PDE4 inhibitor rolipram, enhanced
four-day memory after two training trials, but had
no effect on memory after five training trials.
• The drugs were then injected intraperitoneally
shortly before training to test their ability to
penetrate the blood–brain barrier.
• A number of Helicon’s compounds and rolipram
again enhanced four-day memory after two, but
not five, training trials.
• Stated another way, the PDE4 inhibitors enabled
memory to form following less than half the
normal amount of training.
• Was this effect of the drug specific to one
behavioural task, or did it also work with
other stimuli and sensory modalities?
• addressed by evaluating a second
behavioural task — object recognition —
that relies on a mouse’s natural
exploratory behaviour.
• during training for this task, mice are
presented with two identical novel objects,
which they then explore for some time by
orienting toward, smelling and crawling
over.
• presented at a later time with two different
objects, one of which was presented
previously during training and thus is
‘familiar,’ and the other of which is novel.
• Found that drug-injected mice formed maximal
memory with less-than-normal training
• Thus, PDE4 inhibitors yielded similar effects on
memory formation in two distinctly different
behavioural tasks,
• Evidence that their mechanism of action was on
the memory process per se rather than on a
more nonspecific aspect of a particular
behavioural response.
Further steps in drug development
The blood–brain barrier
• This barrier comprises endothelial cells of brain
capillaries.
• The junctions between these cells are
extraordinarily tight and prevent most classes of
molecules from moving freely from the blood to
the brain.
• This protective function probably evolved to
spare the brain from assault by toxic
substances, viruses and peripheral hormones.
• Unfortunately, it also prevents many
classes of therapeutic drugs from easily
entering the brain.
• the blood–brain barrier is much reduced in
some areas of the brain, such as the
hypothalamus where hormonal signals are
exchanged with the periphery
• To circumvent the blood–brain barrier for normal
metabolism, transport proteins have evolved that
actively carry sugars, amino acids and vitamins
into the brain.
• Attempts have been made to use these transport
proteins to carry drugs, such as glutamatereceptor agonists and antagonists, but the rate
of transport has not been sufficient to deliver
adequate doses of drug to the brain.
• The classic way to penetrate the blood–brain
barrier is to design small heterocyclic organic
molecules.
• These molecules are generally constructed to
have a high degree of lipid solubility (that is,
lipophilicity) and a relatively small size (< 500
daltons).
• Drugs such as rolipram and Helicon’s
compounds were designed with this in mind.
• Such design constraints introduce other
constraints, however, which can include
poor solubility and high affinity for plasma
proteins, such as albumin, leaving only a
small amount of drug free to equilibrate
across the membrane.
• Target specificity (or lack thereof) and high
affinity (< 10 nM) are also desirable, but,
nonetheless, must be balanced against
increased size and the attendant decrease
in brain penetrance.
• CNS drug design constitutes a series of
compromises to balance these various
drug properties.
The P-glycoprotein problem
• Achieving the optimal design to penetrate the
blood–brain barrier does not automatically
ensure that therapeutic levels of drug will reach
the brain.
• There are powerful mechanisms in the more
permeable areas of the brain, such as the
choroid plexus, that actively pump noxious
substances out of the brain.
• One of these pumps, P-glycoprotein, was first
recognized as the multi-drug resistance factor in
cancer chemotherapy
Unique side effects for central
nervous system drugs
• Molecular targets for central nervous
system (CNS) drugs are usually involved
in normal neuronal functions.
• Moreover, they often function in different
types of neurons (excitatory versus
inhibitory, for instance).
• Consequently, CNS drugs can have side
effects that are specific to various brain
functions.
Excitotoxicity
• Some CNS drugs can have excitatory
effects on neuronal activity, which induce
seizures. These may be a direct effect of
the drug or result from indirect activation of
excitatory transmitter (glutamate) release.
Depressant
• CNS drugs can have inhibitory effects on
neuronal function that promote drowsiness
and even unconsciousness and death.
These activities can result from membrane
stabilizing (or local anaesthetic) actions or
from interaction with inhibitory
transmitters, such as GABA (-aminobutyric
acid).
Emesis
• CNS drugs with specific dopamine
receptor activation can cause emesis and
nausea.
Weight gain
• Some nonselective antipsychotic
medications, such as olanzepine, cause
weight gain.
• Receptor knockouts in rodents have
suggested that the 5-HTreceptor may be
involved.
Sexual dysfunction
• Selective serotonin re-uptake inhibitors
(SSRIs) are known to produce sexual
dysfunction.
• Sexual dysfunction has not been reported
as frequently with other classes of reuptake blocker (noradrenaline and
dopamine) and with monoamine oxidase
inhibitors.
• All can function as antidepressants.
• For what diseases might CREB-based
memory enhancer work best?
Stroke rehabilitation
• The extensive, repetitive training that stroke
patients undergo after recovering from the acute
ischaemic trauma is likely to invoke synaptic
plasticity to ‘remodel’ neuronal connections
surrounding the damaged region, thereby
recovering a certain degree of lost function65–
68.
• Co-administration of a CREB-based memory
enhancer and training might yield more rapid
functional recovery (that is, less training will be
required to regain performance).
Alzheimer’s disease
• CREB dependent gene transcription apparently
is not involved in the pathology
(neurodegeneration) associated with AD.
• Moreover, CREB-dependent memory formation
involves the normal chemistry of healthy
neurons.
• Hence, the efficacy of CREB-dependent memory
enhancers might be expected to be lowest when
treating patients with moderate to severe AD.
Treatment of early-stage AD
patients
• Animal models of ß-amyloid pathology
show that memory defects develop well
before any signs of neurodegeneration.
• As neurodegeneration proceeds, however,
individual neurons die more or less
randomly within particular brain regions.
• A kind of mini-stroke.
• Psychological studies of human amnesiacs,
mnemonists and people with exceptional or
normal memory, have demonstrated that
memory is a distinctive cognitive function that
can be measured and studied independently of
other cognitive abilities that humans possess (for
example, reasoning, planning, abstracting,
sequencing, language) or brain functions such
as perception, motivation, emotion or motor
activities.
H.M.
• The patient known as H.M., who has arguably
become the single most studied and quoted
patient in the history of medical research,
developed a severe memory deficit following
surgical removal of portions of his medial
temporal lobes (including the hippocampus).
• H.M.'s severe memory loss, however, was
contrasted by intact cognitive functions such as
perceptual learning, language and reasoning.