AD Dementia - clinical + neurobiology

What is a broad practical definition of dementia?

“A primary and progressive decline of intellect due to structural brain disease to the point that customary social, professional, and recreational activities become compromised”

How does the DSM-V (DSM-5) define Alzheimer’s? What are the criteria you need to meet?

A. evidence of significant cognitive decline from a previous level of performance in one or more areas of cognitive domains (complex attention, executive function, learning and memory, language, perceptual-motor, or social cognition)

• someone is concerned about this person (either the individual themselves, someone who knows them well, or the clinician)

• SUBSTANTIAL impairment in cognitive performance, preferably documented by standardized neuropsychological testing or another quantified clinical assessment

B. the cognitive deficits interfere with independence in everyday activities (so they show up in the person’s daily routine)

C. the cognitive deficits do not occur exclusively in the context of a delirium (a sudden state of confusion)

D. the cognitive deficits are not better explained by another mental disorder (like major depressive disorder, schizophrenia)

Explain the domains under which performance suffers when a person has Alzheimer’s (according to the DSM-V).

• complex attention: increased difficulty in environments with multiple stimuli (information is coming in through multiple senses, which can be overwhelming or hard to deal with), difficulty holding new information in mind

• executive function: unable to perform complex projects, and needs to rely on others to make plans or decisions

• learning and memory: repetitive in conversation, often within the same conversation (brings up the same topics). Cannot keep track of a short list of items when shopping or planning for the day. Requires frequent reminders to orient task in hand.

• language: significant difficulties with expressive or receptive language. Uses general terms (substitutes) for specific nouns or events, like “that thing” and “you know what I mean”. With severe impairment, they may not recall names of closer friends and family.

• perceptual-motor: significant difficulties with previously familiar activities (using tools, driving motor vehicles), navigating in familiar environments/

• social cognition: changes in behavior (shows insensitivity to social standards). Makes decisions without regard to safety. Little insight into these changes

What category of intellectual dysfunction does Alzheimer’s fall under?

Irreversible (not developmental, reversible, or symptomatic)

Does the risk of dementia rise with age? How might that pose a problem?

Yes. This is an issue because we are an aging society, and AD accounts for the majority of all cases of dementia (60-80%). The cost of care will keep rising as long as the number of Americans who have dementia keeps rising

We know that the risk of dementia rises with age. But does it differ by ethnicity, or is there no ethnicity bias whatsoever?

There is an ethnicity bias - there is a higher proportion of those 65 years old or older who are of African and Hispanic descent and have dementia, than Caucasians

Is AD preventable?

This depends. With an early, accurate diagnosis, it is possible to modify the course or trajectory of the disease as well as to ameliorate symptoms.

How do cognitive abilities develop throughout the average person’s lifespan, and when do they peak? At what age does the decline begin?

• cognitive abilities develop through young adulthood, and reach a peak in the 3rd or 4th decade of life

• they are stable through the 5th and 6th decade of life, with SOME decline common in the 7th decline

What areas are declines commonly found in?

• processing speed

• focusing of attention (ability to concentrate)

• reduced mental flexibility and working memory (the ability to hold and manipulate information in your mind)

• reduced learning efficiency and free recall

What is age-associated memory impairment (AAMI)? How is it distinct from Alzheimer’s?

• the patient is at least 50 years old

• the patient has noticed a decline in memory performance

• the patient performs below “normal” levels on a standard memory test

• all other obvious causes of memory decline, except normal aging, have been ruled out

In summary, this is a normal memory decline that comes with aging, and happens to MOST PEOPLE when they get older. It doesn’t interfere with daily life, and the brain is still healthy overall (there is no significant loss of neurons), unlike what happens with Alzheimer’s.

What is MCI (mild cognitive impairment)? What are the two subtypes, and which one is more dangerous in terms of progression to AD? How serious is it relative to AAMI or Alzheimer’s?

• more serious than normal aging, but not dementia yet

• there are memory and thinking complaints/issues, preferably backed up by someone who knows the person well, BUT the person’s cognitive function is generally normal/there is no impairment in functioning (so the person is not demented)

there are two subtypes: amnestic and non-amnestic MCI

• with amnestic MCI, memory problems (as hinted by the name) are the biggest issue - like forgetting appointments or names

• there is a larger risk of progressing/converting to AD (a small percentage of people actually do).

• However, around 40% of people with MCI revert to normal per year.

Talk about the progression of someone from normal, age-associated memory decline to MCI to dementia (most of which is caused by AD).

We can measure this progression via memory tests, though the reliability of this is imperfect if someone is able to take the test multiple times and get used to its content and format. Also, a score on a cognitive test that looks like mild cognitive impairment (MCI) might actually reflect a person’s normal cognitive ability - not a TRUE decline. When we do another test, we might see the score fall in the normal range.

However, through looking at memory tests, we can track the progression from age-associated memory decline (with an observed score of 2) to MCI (with an observed score of 1) and then to dementia.

Though Alzheimer’s is responsible for most cases of dementia, what are some other causes?

• Alzheimer's disease (70%)

• Dementia with Lewy bodies (10%) - also result from clumps of a misfolded protein. The symptoms of this disease are somewhere between Parkinson’s and dementia

• Frontotemporal dementia (10%) - caused by Pick bodies (protein aggregates) in the frontal and temporal lobes, and the main difference is that behavioral and emotional changes result before changes in memory

Talk about cognitive trajectories - the continuum of Alzheimer’s disease as seen through the preclinical and symptomatic stages. What treatment model does this support?

• preclinical stage (before symptoms show up, which is known as the asymptomatic stage) - substrate (the amyloid plaques and tau tangles) is still responsive (to proteins like BDNF that promote neuronal survival and to therapeutic inventions, like drugs)

• the aging (symptomatic) stage, where the substrate becomes unresponsive (since dead neurons can’t be revived and damage is too extensive to recover function). In terms of gross cortical changes, we see massive shrinkages of gyri (cortices) and widening of sulci (grooves), in addition to enlargement of ventricles due to tissue loss.

This supports the early intervention model of Alzheimer’s (where diagnosing and intervening is crucial to stabilizing cognitive function)

What does quality of life in AD look like?

There’s a steep decline as clinical symptoms (within the domains outlined in the DSM-V criteria, such as cognitive impairment, quality of life, social dependence, and motor abnormalities) worsen.

What are the following things that lead to neuronal death?

• apoptosis, obviously

• excitotoxicity (which can trigger apoptosis via Ca2+ influx, which can open the mitocondrial PTP and lead to the leakage of cyt-c in the cell, which activates caspases)

• mitochondrial dysfunction

• calcium dynamics

• inflammation

What is the underlying neuropathology behind Alzheimer’s disease? Name two of the protein forms that contribute to the pathology.

plaques - agglomerations of a pathogenic form of amyloid beta (a peptide that normally has a lot of different functions (activation of kinases, transport of cholesterol)

(neurofibrillary) tangles - interwoven strands of HYPERphosphorylated tau proteins, which interfere with internal neuronal transport mechanisms

plaques are extracellular, tangles are intracellular

How does the pathology of AD spread throughout the brain?

Both amyloid beta and tangles spread from the limbic system/basal forebrain to the hippocampus to the neocortex (from the center/underside of the brain to the outer surfaces).

AD is strongly associated with the loss of cholinergic neurons (particularly those in the basal forebrain), which project to glutaminergic neurons in cortex and hippocampus. It also targets the pyramidal neurons in the cortex and hippocampus (which project to other regions of the brain).

When gene expression was measured among different subtypes of excitatory neurons, it was found that there were differences in the patterns of gene expression. What were the findings?

There were two neuronal subtypes:

• vulnerable excitatory neurons: large pyramidal neurons in the cortex and hippocampus (which were more vulnerable to tau pathology and faced changes in gene expression, like downregulation of synaptic genes + upregulation of stress response/apoptotic pathway genes)

• resilient excitatory neurons (with more limited connectivity - which exhibit higher expression of genes relating to synaptic plasticity)

How do amyloid plaques and neurofibrillary tangles contribute to dysfunction?

• interfere with synaptic contacts (which can impair long-term potentiation and synaptic strength)

• affects axonal transport (causing “traffic jams” when motor proteins, which try to transport vesicles and proteins, are disrupted)

• regulate membrane dynamics (disrupting membrane potential)

• calcium homeostasis

What is the intermediate between misfolded proteins/abnormal conformations to protein aggregates?

we go from proteins with abnormal conformations —> small (but still soluble) oligomers that are capable of binding to synaptic receptors, like NMDA, to trigger calcium influx —> (large, insoluble) aggregates or fibrils (which interfere with cellular and axonal transport and glial function)

What affects Abeta (plaque) accumulation?

• reduced clearance (being less effective at getting rid of it) —> via impairing integrity of the blood-brain barrier, which affects the functioning of apolipoprotein (ApoE) moving cholesterol and phospholipids through the bloodstream/CSF. ***in its harmful isoform, ApoE also binds to amyloid beta peptides directly and can promote amyloid beta aggregation, which makes clearance harder. Future treatments might target clearance of plaques***

• increased production (via mutations that can be inherited/passed down to offspring)

• reduced degradation (via changes to/impairing the ubiquitin proteasome pathway)

• inflammation (feed forward process - the initial event triggers changes that reinforce or worsen the original problem)

What are the genetic factors in AD - how do they differ between factors that CAUSE AD, and factors that simply increase your risk of developing it?

cause AD:

• autosomal DOMINANT genes (though this is a very small minority of cases) —> cause AD (rather than just increasing the risk). If you have even one copy of the gene, you almost certainly will develop AD.

  • amyloid precursor protein —> mutations in APP produce amyloid beta 42, which is a TOXIC and STICKY peptide that is longer and more hydrophobic (meaning that it has a greater tendency to aggregate into plaques)

  • presenilin 1 (part of the gamma secretase complex, which cleaves APP) —> also leads to production of amyloid beta 42

  • presenilin 2 (part of the gamma secretase complex)

increase the risk of having AD:

• ApoE4 allele - increases risk significantly and lowers the age of onset, but does not guarantee AD. One copy of the allele (heterozygous) —> 3 times risk, two copies (homozygous) —> 15 times the risk of those who are wild-type for both

How can someone carry a PSEN2 (presenilin 2) mutation and not have cognitive decline?

• incomplete penetrance (not every person with the mutation develops the disease) —> may be due to protective factors elsewhere in the genome

What is ApoE, what type of cells, and how many alleles exist?

ApoE is a protein that is found in the astrocytes in the CNS (which maintain the blood-brain barrier) and is a protein that is important for cholesterol transport across the BBB (blood-brain barrier)

• ApoE2 (protective)

• ApoE3 (neutral, most abundant)

• ApoE4 (harmful, about 3x the risk of developing AD AND cardiovascular disease) —> shown by a study looking to test for an AD susceptibility (risk-increasing) gene, comparing three versions of the allele to each other using age (x-axis) compared to risk of developing AD (y-axis)

What are the treatment options for AD? Rank them as they correspond to the increasing severity of the disease (as it progresses)

• neuropsychiatric symptoms - psychosis medication

• cognition enhancer - 5-HT (serotonin) antagonists

• disease-modifying therapy (DMT): oral drugs

• combination therapy: DMT + cognition enhancer

What do cholineresterase inhibitors like memantine do, and why might they be a treatment possibility for Alzheimer’s?

Cholinesterase is an enzyme that breaks down acetylcholine, which is important because AD targets, and causes the degeneration of, cholinergic neurons in the basal forebrain - ULTIMATELY leading to less ACh being released in the synapse.

So we want to reduce the breakdown of ACh in the synaptic cleft so as much of it remains as possible, which increases neuronal signaling/communication.

Do we have a cure or any diease-modifying therapies for Alzheimer’s?

No. We just have symptomatic treatments that are minimally effective.

What are the biomarkers for preclinical AD, in order of appearance?

during the preclinical/age-appropriate cognitive decline stage (reach a high or peak during MCI or mild cognitive impairment):

• amyloid B accumulation (PET imaging)

• Synaptic dysfunction (PET imaging)

• Tau mediated neuronal injury

mostly appear either right when the preclinical stage ends, or where MCI begins, and reach a peak in the dementia stage:

• brain structure (as seen within an MRI)

• cognition

present during dementia:

• clinical function (requires patient care)

What are the protective factors against Alzheimer’s?

• high education level

• hobbies

• aerobic exercise/strength training

• cholesterol-lowering strategies (statin therapy)

• absence of hypertension and diabetes

• ApoE2 (positive, opposite of risk-causing) allele

• amyloid beta clearance

What are the four active research areas in AD?Alzheimer’s?

1. Cholinergic neurons

   • Nerve growth factor/NGF - retrograde (from axon terminal, where it is received from target areas in the cortex and hippocampus, to cell bodies in the basal ganglia) transport of NGF is CRITICAL for basal forebrain cholinergic neurons (largely found in the ventral part of the basal forebrain). !! Even if NGF is present, if the retrograde transport mechanism is affected, these neurons’ survival will be affected.

   • Brain-derived neurotrophic factor (BDNF) - supports synaptic plasticity and long-term memory (LTP) of glutaminergic neurons (which have NMDA receptors), which is decreased in AD.

2. Amyloid beta - fragment of the transmembrane protein APP that is cleaved twice (once by beta secretase, and once by gamma secretase - this is known as the toxic pathway). It has normal functions but as a (toxic) extracellular protein aggregate, it forms plaques that disrupt synaptic signaling and impair LTP of synapses, in addition to causing activation of microglia (brain’s immune system) —> triggering chronic inflammation

3. Tau - hyperphosphorylated tau proteins form intracellular tangles, present a STRONGER correlation with cognitive decline than amyloid beta aggregates (plaques) do. Since the purpose of the tau protein is to stabilize the cytoskeleton and maintain the structural integrity of axons (which are long), tau tangles can cause cytoskeletal collapse. Also, under normal conditions, tau proteins associate with microtubules (which are the highway that motor proteins walk on to transport vesicles, proteins, etc. —> without the microtubule stabilization provided by tau proteins, we lose the ability to transport proteins like NGF.

4. ApoE (apolipoprotein) - the brain’s lipid shuttle, which helps deliver cholesterol and lipids to neurons (important for the integrity of their membranes). It also binds to amyloid beta and helps clear it away from the brain.

   • the ApoE4 isoform is worse at clearing amyloid beta than ApoE2 or ApoE3. So ApoE2 = protective isoform, while ApoE4 increases AD risk and ApoE3 has no effect.

What is the cholinergic hypothesis for AD? Hint: it has to do with NGF

1. basal forebrain cholinergic death correlates with the severity of disease (as in, the progression of AD is linked to how many cholinergic neurons in the basal forebrain have died)

2. Though NGF is important for neuronal development everywhere in the brain, CBF neurons ESPECIALLY depend on NGF for their survival and maintenance (known as trophic support)

3. An inability to obtain or use NGF underlies the death of these CBF neurons.

How does the process of secreting NGF (done by the glutaminergic neurons) and receiving it at the cholinergic neuron cell bodies work?

1. glutaminergic target neurons secrete NGF

2. NGF diffuses across the synapse to the cholinergic neuron axon terminals, binding to TrK A and p75 receptors

3. An NGF-receptor complex (meaning BOTH NGF and its target receptors) is transported to the cell bodies of the glutaminergic neurons via retrograde transport on the microtubule highway

4. Once at the cell body, they trigger/activate the expression of pro-survival + growth genes.

Talk about the p75 receptor - under what conditions might it be linked to apoptosis? What happens to the balance between Trk A and p75? Which receptor is high affinity?

• co-activation with TrK A in the cortex and CBF neurons (meaning that NGF binds to BOTH Trk A and p75 receptors) —> can enhance the binding of NGF to Trk A and/or signaling produced by Trk A and is ultimately pro-survival

• activation alone (by an immature form of NGF) —> can activate apoptotic pathways via caspases

So under Alzheimer’s disease conditions, p75 levels are upregulated (boosted) while TrK A levels go down. The balance TIPS in favor of apoptosis, especially because apoptosis signaling can happen LOCALLY (within the axon terminal, since caspases are still present there) without the need to travel to the soma (via the microtubules, which are impaired under Alzheimer’s via the hyperphosphorylated tau proteins - meaning that we also see less of the NGF-Trk A complex, which is pro-survival, being transported to the soma)

Trk A is high affinity (for NGF) while p75 is not - since p75 is a generic receptor that can bind to all neurotrophins (ex. BDNF).

Is there an age-related decline in NGF transport?

Yes.

What is an easy tell to see if a cholinergic neuron cell body is receiving NGF?

If it has a big cell body! Small cell body = it is not getting NGF, hinting at impairing of axonal transport.

What is ChAT?

ChAT = choline acetyltransferase, involved in the production of acetylcholine (which is obviously expressed by the neurons of the basal forebrain). Obviously, cholinergic neurons will express ChAT.

We know that cell body size is linked to NGF being present in the cell body by studies looking at the size of neurons that express both (double-labeled with) NGF and ChAT (big), compared to just expression of ChAT (small). But for both groups there was a reduction in cell size for aged cells compared to young cells (possibly linked to decreased ChAT/NGF expression)?

From the same study, we also found that the number of ChAT (cholinergic) neurons that were labeled with NGF DECREASED in aged neurons compared to young neurons - supporting less NGF expression or uptake as the cell ages.

What is the amyloid hypothesis of AD?

• amyloid beta is the primary component of senile plaques

• senile plaques disrupt neuronal function (since oligomers are clustered around synapses and lead to communication deficits between cells) + lead to death

• mutations in APP (amyloid precursor protein) cause familial AD (a heritable form) —> since it is an autosomal dominant disease, this provides a high chance you will end up getting it

We know that mutations in APP (amyloid precursor protein) lead to APP mismetabolism (cleavage into amyloid beta), but is an increase in amyloid beta aggregation related to an increase in Tau protein? Which pathology is correlated better with cognitive decline, and which pathology “sets the stage”?

Yes, since amyloid beta oligomers (soluble forms of amyloid beta) aggregate outside of the cell and trigger the activation of kinases, which add phosphate groups to proteins.

As you might guess, they add extra phosphate groups to tau proteins (hyperphosphorylating them, and leading to the development of tau tangles). So the accumulation of amyloid beta often precedes tau pathology.

Amyloid beta sets the stage, but tau tangles are what are believed to drive cognitive decline/neurodegeneration.

Talk about APP processing, and the three secretases. Which are good, neutral, and bad? What do PS1 and PS2 do?

Alpha secretase - part of the non-amyloidogenic (safe) pathway. Good because it cuts in the middle of the APP protein.

Beta secretase - first step on the amyloidogenic pathway (which means that it sets us on the path to create amyloid beta)

Gamma secretase - neutral, because it can be behind a step on either pathway

• safe/non-amyloidogenic pathway: works after alpha secretase cleaves the APP protein in half to make a non-toxic peptide(s).

• toxic/amyloidogenic pathway: works after beta secretase

Because gamma secretase also has other functions in the cell, and has a role in the non-amyloidogenic pathway, we don’t want to inhibit it. We want to inhibit beta secretase.

PS1 and PS2 = encoded by PSEN1 and PSEN2

• mutations in PSEN1 and 2 = lead to early onset Alzheimer’s

• PS1 and PS2 —> act as the “blades” to the scissors of gamma secretase

What is the fate of amyloid beta after it is created?

• can be released extracellularly (form oligomers and protein aggregates)

• chewed up/degraded by microglia or taken up by receptors (aided by ApoE)

What does APP do (since, if we get rid of it, we could lose some of its more positive functions)?

• axonal transport via the motor protein kinesin particularly of complexes containing its proteases (proteins that cleave other proteins), beta and gamma secretase

• cell adhesion (maintaining synapses/”linking” dendrites to axon terminals

  • synapse formation?

  • development

• cell growth

  • development

• synaptic plasticity

  • LTP and dendrite complexity in the hippocampus (HUGE site for learning and memory, and also dendritic growth)

How does amyloid beta cause a shift in caspase-3 activity (known as the caspase-3 bifurcation switch) with age?

Contributing factors to active caspase-3 include neuronal activity, amyloid beta levels/aggregation, and ROS (reactive oxygen species, like free radicals). Also, caspase-3 can have two functions: modulation (influencing) of synaptic plasticity, or cell death.

In a young cell, we see high neuronal activity, but LOW levels of amyloid beta and reactive oxygen species - producing the route of synaptic plasticity modulation.

In an aged cell, we see high levels of ALL three. This leads us towards the cell death pathway.

Beta amyloid toxicity varies by brain age —> if it activates caspase-3 and we see synaptic modulation happening instead of cell death, we can say that amyloid beta plays a beneficial role when young but a toxic (cell death) role when old.

Talk about the graph looking at the caspase-3 bifurcation switch, with three axes.

we had three axes = proximity to synapse, signal intensity, and duration of activation (length).

Longer activation and lower proximity to synapse, as well as an increase in signal intensity = cell death (unfavorable, toxic conditions, like what we see when a person is aging)

Higher proximity to synapse but lower signal intensity and activation length = synaptic plasticity modulation (favorable conditions, like when a person is young

Talk about the APP knockout mice. How do they differ from the control/wild-type mice? What did results of the study show?

These mice are viable (able to be born), instinguishable from wild-type mice in appearance and locomotion (so the differences would be a lot more subtle). They do have some defects, though

• defect in passive avoidance learning with age (when mice learn to avoid an area where they previously received a mild shock) —> may link to LTP defects

• defect in retinal wiring (but no effect on visual acuity/sharpness of vision)

• defect in limb strength (decreased grip strength or coordination)

• The results of the study likely point to a toxic gain of function - so not a failure of APP’s normal role (loss of function), but a NEW, harmful effect caused by the mutation.

What are some models of Alzheimer’s disease? Which model leads to the development of amyloid beta plaques, loss of synapses, spatial memory, and gliosis (inflammatory + repair response of the brain)

• familial mutation mice (heritable version of AD)

  • APP

  • PS1 (genes that encode the scissors of gamma secretase)

  • PS2 (genes that encode scissors)

• combination mice

- PS1 + APP

We see Abeta plaques, as well as the other symptoms (like loss of synapses and development of gliosis/inflammation) in the combination mice.

Talk about the triple transgenic mice, which are the most commonly used model because of how pathology-driven they are in comparison to other models (which means that they involve multiple of the core features of Alzheimer’s disease, NOT just the amyloid beta pathology). What human mutations are these mice homozygous for, and what do these mutations affect?

• APP mutation that leads to overproduction of amyloid beta

• presenilin-1 mutation (alters the activity of PS1, which is part of the gamma secretase complex that leads to AB being produced

• Tau mutation that leads to hyperphosphorylation of tau protein and formation of tangles

The reason why these mice are more pathology-driven is because they have BOTH the amyloid beta and the tau pathology, and studies with them show that the presence of amyloid beta drives tau pathology.

What do studies using the triple transgenic mice show?

• Abeta and tau pathology closely follows human AD in terms of timing and location (we see the formation of Abeta aggregates first, followed by the formation of Tau tangles thanks to kinase activation caused by these aggregates)

• Deficits in LTP come BEFORE Abeta deposits

• cognitive decline begins at 4 months (though the human model is not that aggressive/doesn’t have such an early onset of the pathology, raising questions about the degree of applicability of the model to humans)

• Abeta immunotherapy (using the body’s own immune system) DELAYS the decline of memory

What is the difference between amyloid beta 40 and amyloid beta 42?

the number at the end corresponds to how many amino acids are present - amyloid beta 42 has 42 amino acids in total, which makes it MORE hydrophobic and prone to aggregation than 40. This is because amyloid beta peptides come from the C-terminus end of APP (which is MORE hydrophobic than the N-terminus, and more prone to forming beta sheet-rich structures where the beta sheets stack, stabilized by hydrogen bonding, to create insoluble aggregates), and the longer the peptide, the MORE of the C-terminus that is included.

Talk about what pathology looks like in the 3x mouse model of AD. How does amyloid beta spread in the brain (what regions does it start in?) and what do we use to detect this pathology? Do we see an LTP defect?

• deposition of amyloid beta plaques occurs in the neocortex and progresses to the hippocampus (this is the amyloid-centric pathology, which spreads from the outer brain inwards, in contrast to the tau-centric pathology of spreading from the cholinergic basal forebrain nuclei to the hippocampus to the neocortex).

• We can use antibodies that are specifically designed to bind to amyloid beta 42 to detect its spread through the brain. These antibodies pick up on extracellular amyloid beta plaques AS WELL AS immunoreactivity (detection of the plaques) within neurons, which may either indicate a reaction to the intracellular tau tangles, or might be a response to the detection of any amyloid beta found intracellularly (before the amyloid beta leaves the cell - ex. through exocytosis, where it can leave the cell via vesicles and then form extracellular plaques).

• We do see an LTP defect (shorter excitatory postsynaptic response) in the 3x transgenic mice as compared to control mice (no mutations) and mice with a PS1 mutation (meaning that the gene that codes for PS1, known as PSEN1, has been mutated and the resulting gamma secretase complex that the protein PS1 is part of will produce more amyloid beta 42 (TOXIC) over amyloid beta 40. There’s also a shorter excitatory postsynaptic response in the 3x mice than 2x mice, indicating the DETRIMENTAL effect of these mutations.

Can animal models be used to develop better treatments?

Yes!

• animal models allow TARGETED, hypothesis-driven testing of drugs based on what we know about how the disease works and what mechanisms it progresses by. Basically, we can manipulate specific genes or pathways in animals (ex. mutating the PSEN1 gene that codes for the PS1 protein, which forms part of the gamma secretase complex), and observe the effects/see how those pathways affect disease and treatment outcomes.

• short life span of rodents - since rodents only live a few years, you can study the progression of disease, and the efficacy of treatment, much faster than in humans.

• less complex brain anatomy - since rodent brains are simpler than humans, with fewer neurons and circuits, isolating what we specifically want to experiment on is less difficult.

  • One caveat is that, while the structure of neuronal cells is more evolutionarily conserved between species, glial cells differ more significantly between mice and humans. Human glial cells = larger, more complex dendritic trees.

  • We can overcome this by implanting human glial cells in mice models, which creates more “humanized” models with pathologies more applicable to us.

• animal models also allow us to compare current treatments (what we are doing now) with new treatments to determine the efficacy of each.

What are BACE inhibitors, and what is the reasoning for proposing to use them in the treatment of Alzheimer’s? Do they actually work in terms of reducing the amount of amyloid beta?

• assumption: Abeta is toxic to neurons, promotes inflammation, and likely affects tau pathology (via activating kinases that phosphorylate tau and lead to an inability for it to carry out its function of microtubule stabilization, affecting the integrity of the cytoskeleton and transport of proteins, etc. from axon terminal to soma and vice versa)

  • so, reducing abeta production should benefit not only symptoms, but stop disease progression via feed-forward cycles (accumulation of amyloid beta creates a stressful, inflammatory environment that indirectly enhances beta secretase or BACE1 activity, which leads to more cleavage of APP to amyloid beta)

  • rather than target gamma secretase, which plays a role in the non-amyloidogenic pathway, we should target the precursor enzyme (beta secretase, which (in the context of APP) is exclusively active during the amyloidogenic pathway. If we target BACE1, this precursor step, we won’t end up with the final toxic product (amyloid beta)

• when four groups of mice (one with the Swedish APP mutation that makes APP a better substrate for BACE1, and one wild-type group - with each group being split into sub-groups where one was given a solution with no BACE1 inhibitor and one WAS given BACE1 inhibitor) we see much lower levels of amyloid beta in the inhibitor groups for both wild-type and Swedish APP mice.

Is there reduced pathology in BACE-1 knockout mice?

Yes! Basically, this is where the gene (and thus the resulting protein product) that codes for beta secretase has been knocked out, and we see reduced production of Abeta as a result. We also see (in memory tests, where mice explored an environment twice and both durations were compared to each other - it was expected that if the mice remembered they had explored this environment before, they would spend less time doing so) that Swedish APP mice groups that were treated with BACE1 inhibitor show significant improvement in memory compared to Swedish APP mice without the inhibitor - and that this reduction was almost equivalent to the wild-type mice.

Talk about taupathy (tau pathology) in AD - what is the role of normal tau proteins, and what is a factor that contributes to their hyperphosphorylation besides kinase activity?

• healthy microtubules use tau proteins to stabilize them (as in, tau proteins are associated with microtubule proteins). The tau proteins keep the microtubules in a tight conformation.

• When phosphorylated, the tau proteins lose their affinity for the microtubules and instead “prefer” to associate with each other/bind to OTHER tau proteins, which creates tau tangles and leads to the loosening/unraveling of the tight microtubule conformations.

• This has an effect on the structural integrities of both the cytoskeleton and the axon terminals (decimating axonal transport).

• one theory as to why tau proteins are allowed to remain phosphorylated (as in, a factor besides just activation of kinases that phosphorylate tau proteins) is the reduced availability of phosphatases, which are enzymes that remove phosphate groups. Usually kinases and phosphatases are involved in a tug-of-war, but if the balance tilts overwhelmingly in favor of kinases…we can end up with taupathy.

Talk about how cAMP-calcium signaling is linked to hyperphosphorylation of tau, as seen in primate brains.

• in the primate dorsolateral prefrontal cortex, mGluR3 receptors (metabotrophic glutamine receptors - not the same thing as NMDA receptors, which are responsible for learning and memory and directly open the Ca2+ ion channels. mGluR3 receptors activate second messengers like cAP) regulate cAMP-Ca2+ signaling.

• as the brain ages, glial cells produce more GCP2, which is an enzyme that breaks down NAAG. Basically, NAAG inhibits cAMP production, which lowers Ca2+ influx and protects against excitotoxicity. If we have higher GCP2, we have less regulation performed by the mGluR3 receptors (less inhibition against cAMP production and greater Ca2+ influx as a result, since cAMP activates PKAs - protein kinases - that phosphorylate proteins that work on calcium channels).

• This leads to the activation of tau kinases, which phosphorylate tau + lead to the development of Alzheimer’s.

What do studies find that limit the phosphorylation of tau but involve an excess of amyloid beta (likely through increasing production)?

• we see evidence that tau may contribute more to cognitive decline/pathology - synaptic and neural damage is also less severe

• this shows that amyloid beta doesn’t fully drive the cognitive decline (even when levels are high) - phosphorylated tau is needed to cause major dysfunction

Talk about the ApoE4 allele (ApoE plays a role in the clearance of amyloid beta, but it comes in different isoforms).

ApoE4 is the negative (risk-increasing) isoform - so if you have 1-2 alleles of it, you will have reduced amyloid beta clearance and increased amyloid beta aggregation as a result (a loss of physiological function). You will also see an increase in tau tangles/neurodegeneration and an increase in brain atrophy.

Also, ApoE is involved in lipid (cholesterol) transport - since APP is a transmembrane protein, too much cholesterol (in the membrane) means that APP is more likely to localize there, AS WELL AS gamma secretase. This is problematic because gamma secretase cuts inside of the cell membrane - so overall, the amyloidogenic pathway becomes more likely and the amount of amyloid beta in the cell is increased.

What is the DNA hypothesis that explains the difference in the degree of harm caused by the ApoE2 and E4 alleles? Talk about DHA as well, which is a special lipid.

the E2 and E4 alleles differ by just one or two nucleotides, which result in amino acid changes in just two regions of the final ApoE protein. ApoE4 is less efficient at transporting lipids - which not only include cholesterol (incorporated in membranes and harmful where APP and gamma secretase localization is concerned), but the omega-3 fatty acid DHA.

DHA is important for development and synaptic plasticity. ApoE2 is better at transporting DHA to neurons, reducing AD pathogenesis and risk, than ApoE4. This increases the risk of developing Alzheimer’s.

How does neuroinflammation influence amyloid beta accumulation (specifically the Abeta 42 isoform, which is more prone to aggregation and neurotoxicity).

mouse models were genetically edited to express:

• pro-inflammatory cytokines

• anti-inflammatory cytokines

• control group (no editing)

Then all groups were treated with amyloid beta 42. The levels of inflammation were measured by looking at the levels of mRNA expression for the cytokines. Pro-inflammatory cytokines (ex. IL-6) were found to be higher under AD conditions (given via Abeta treatment) - which likely contributed to neuroinflammation and neuronal damage.

Talk about how inflammation feeds forward in AD pathology. How does the presence of beta amyloid indirectly enhance the activity of BACE1 (beta secretase)?

• presence of beta amyloid leads to activation of microglia and astrocytes - microglia form the brain’s immune system/response

• inflammatory mediators (messenger molecules that amplify or carry the alarm) are released

• NF-KB (which can be activated both by amyloid beta and by cytokines that are part of the signal cascade that amyloid beta triggers) is a transcription/injury response factor that can alter how much inflammatory genes are expressed. It dials this UP.

  • the activity of NF-KB can also be triggered by oxidative stress, ischemia (blood supply/restriction issue), and traumatic brain injury

• Inflammatory conditions activate BACE1, increasing APP amyloidogenic cleavage + levels of Abeta

So we end up seeing the chronic inflammation that contributes to reactive glial cells (which kill neurons and contribute to neurodegeneration and cognitive decline)

What does it mean for astrocytes and neurons to be “implicated separately”?

In some cases, neurons and astrocytes are implicated separately - meaning that since gene expression varies by cell type, there are some mutations that only affect glial cells like astrocytes and some mutations that only affect neurons.

Talk about astrocytic glt-1 in Alzheimer’s disease (the 3x transgenic mouse model)

Remember that the job of astrocytes is to help create the “best” environment for neurons to remain functional in. Astrocytic GLT-1 (the main glutamate transporter responsible for clearing extracellular glutamate) is generally DOWNREGULATED or dysfunctional in Alzheimer’s disease conditions (shown by a study where there was a reduction in expression levels over time)

• a reduction in GLT-1 leads to impaired glutamate clearance, which leads to excess extracellular levels of glutamate (which can lead to synaptic dysfunction via overactivation of NMDA receptors, leading to excitotoxicity via Ca2+ influx).

• Ceftiraxone (an antiobiotic) increases GLT-1 levels by upregulating the expression of GLT-, which improves behavior performance (ex. the recognition index, a measure of memory, shows increased performance on Ceftiraxone). There was also improved performance (measured by escape latency, another measure of memory when mice who were exposed to cues were timed to see how long it took to swim in a water maze, find an invisible platform, and stand up on it).

• Ceftiraxone also acts on soluble Abeta + tau levels, which is really promising.

How can enhancing microglia activity in terms of clearing away Abeta plaque accumulation be beneficial?

Microglia are the brain’s immune cells that help clear debris, including Abeta plaques.

• around plaques, brain tissue becomes stiffer thanks to protein aggregation and local inflammation

• Piezo1 is a mechanosensitive (pressure-sensitive) ion channel. So when it feels this tissue stiffness, it is activated and signals microglia/enhances their phagocytic ability

• Yoda1 is a drug that activates the piezo 1 channel, increasing its ability to signal to the microglia that there are plaques that need to be broken up/toxic waste that needs to be cleared.

• this leads to an increase in learning and memory

What are some emerging methods in AD research?

• epigenetic analysis (acetylation/methylation)

• gene expression studies (on the level of mRNA levels)

• protein analysis (via mass spectrometry, which identifies proteins based on a mass to charge ratio, or using antibodies/Western blots)

What are the advantages of single-gene expression profiling, like what we do with qPCR?

• highly specific and targeted (measuring one gene of interest with a high degree of sensitivity)

• can be tissue or cell-type specific (useful when there is a difference between types of excitatory neurons - for example, in terms of looking at excitotoxicity, we have vulnerable neurons and resilient neurons).

  • neurons that express proteins like MEF2 (which is a transcription factor involved in the upregulation of genes relating to neuronal survival and synaptic plasticity) are more resilient. So more active MEF2 = more resiliency.

  • neurons that express higher ApoE(!) and MHC1 expression (aka compensatory mechanisms for increased risks of amyloid beta production/plaque buildup) are more/selectively vulnerable to AD. **This is an especially relevant finding because astrocytes are the MAIN SOURCE of ApoE in the brain, and under Alzheimer’s/chronic inflammatory conditions, we see the astrocytes express it less while vulnerable neurons try to express it more, even though ApoE expression isn’t their “job”, as a compensatory mechanism.

single-gene expression profiling might also reveal differences within glial cells. For example, APOE (involved in clearance of amyloid beta) is downregulated/less expressed in AD astrocytes but upregulated/more expressed in microglia.

(big takeaway - in AD conditions ApoE is upregulated in neurons, and downregulated in astrocytes)

How might an HDAC inhibitor serve as a treatment for AD? Think about how can we link histone deacetylation (the role of HDACs) to tau phosphorylation.

We find that HDAF inhibitors might lead to preservation of learning and memory or restoration of lost memories (in mouse models that have been engineered to exhibit AD pathology/neurodegeneration).

HDACs are responsible for histone deacetylation - which leads to a downregulation of target genes. HDACs are often activated under stressful or inflammatory conditions (like what we see with AD), and often lead to DECREASED EXPRESSION of genes that are important for synaptic plasticity, neuroprotection….and even the phosphatases that normally dephosphorylate tau proteins!

HDAC3 activity/overexpression of HDAC indirectly shifts the balance in favor of kinases (and phosphorylated tau in comparison to dephosphorylated tau), which is bad.

• However, when we use an inhibitor to decrease/reduce HDAC activity, we end up seeing an increased amount of dephosphorylated tau. So less histone deacetylation = less tau phosphorylation

Does the HDAC-3 inhibitor also affect amyloid beta?

Yes. It’s kind of mixed because it while it does decrease amyloid beta 42 (the more toxic version), it leads to an increase in amyloid beta 40. But overall the benefit of decreasing 42 outweighs that. We also saw an increase in behavioral performance (the barnes maze test and barnes maze errors) for mice treated with HDAC-3 inhibitors.