Williamson Alzheimer's

Question 1

What are amyloid diseases?

Answer 1

These are diseases where a naturally occurring protein (which normally is globular and has a well-understood function) unfolds, misfolds, aggregates, and collects in fibrous plaques.

Question 2

What are amyloid plagues always formed of?

Answer 2

Beta-sheet protein (whatever the original globular protein looked like), and in an X-ray beam they all look similar – they form a ‘cross’ indicating repeated structure every 4.7 Å in one direction and 10 Å at right angles. This is caused by parallel -strands, running at right angles to the fibre direction – “cross-”.

Question 3

What are examples of amyloid diseases?

Answer 3

1. Systemic amyloid disease: Body produces excessive amounts of antibody light chains, which accumulate as amyloid deposits in different tissues (particularly kidneys).
   Also builds up in heart and causes heart failure.
2. Acute phase proteins are proteins that are produced by the body to deal with inflammation – eg lots of innate immunity proteins.
   One of these is serum amyloid A, which recruits immune cells and transports cholesterol to the liver. It is upregulated in chronic inflammations such as rheumatoid arthritis and therefore builds up. Can produce amyloid deposits in kidneys.
3. Dialysis is used when kidneys fail. The normal function of the kidney is to filter the blood and remove harmful stuff. Dialysis uses external dialysis filters.
   It does not remove proteins (too big for the dialysis filters) so proteins build up with long-term dialysis. One of these is 2 microglobulin, a component of MHC class I. It builds up in organs and joints: dialysis-related amyloidogenesis. After it was recognised as a problem of dialysis, it was treated and is no longer a problem. Interesting example of an iatrogenic disease.
4. Type 2 diabetes is a consequence of insulin resistance; one response of the body is to increase expression of insulin.
   A hormone called amylin or islet amyloid polypeptide (IAPP) is expressed at the same time as insulin, and about 100x more. The body therefore makes lots of IAPP, and diabetes patients tend to get amyloid deposits of IAPP in the islet cells.

Question 4

What are the symptoms of Alzheimer’s?

Answer 4

First symptoms are loss of short-term memory. Later symptoms include disorientation, mood swings, delusions and apathy, leading to withdrawal. Average life expectancy after diagnosis is 3-9 years.

Question 5

Why are the plagues called amyloid?

Answer 5

When first seen (1850s) they stained with iodine in the same way that starch does so they were called amyloid (starch-like).
Actually they have nothing to do with starch and are made of protein.

Question 6

What is the protein called that plagues are made of in Alzheimer’s?

Answer 6

In 1984, the protein was extracted from plaques, purified and sequenced. It is called amyloid-beta (A) and is 39-43 residues long. The plaques surround the neuron and are outside it. Inside neurons, you can often see neurofibrillary tangles. These turn out to be a hyper phosphorylated form of the protein tau, which binds to microtubules and stabilises them.

Question 7

What is A-Beta protein made from?

Answer 7

ABeta is cut out of a much longer protein called amyloid precursor protein, APP. The function of APP is not known. Something to do with synaptic function but not clear. The protease Alpha-secretase cuts in the middle of the ABeta sequence and therefore lowers the amount of ABeta. Probably most APP is cut by Alpha-secretase. It is part of a large family of proteases called ADAMs.

However, if APP is cut by Beta-secretase (BACE), it creates an extracellular fragment and a transmembrane fragment. The transmembrane fragment is not then cut by Beta-secretase, but is cut by a Gamma-secretase also called presenilin. The presenilin genes were discovered by linkage studies using mutations in familial Alzheimer’s. Beta-secretase produces a peptide that is either 40 or 42 residues long: ABeta(1-40) or A(1-42). The 1-42 is more prone to aggregation and more toxic.

Question 8

What is the genetic basis of Alzheimer’s?

Answer 8

Most cases of Alzheimer’s have no obvious genetic element. However some do, especially early onset Alzheimer’s (before age 65). This is caused mainly by mutations in the gene for APP or the genes for presenilins 1 and 2. This led to the amyloid cascade hypothesis for Alzheimer’s: that it is caused by overproduction of ABeta, which collects into plaques; and the plaques are in some way toxic. An obvious consequence is that reduction in the amount of ABeta could be a cure for the disease.

Question 9

What genetic basis for Alzheimer’s is linked with Downs syndrome?

Answer 9

Further support is that the APP gene is on chromosome 21. Trisomy of chromosome 21 causes Down syndrome; and downs patients typically all have early onset Alzheimer’s (age 40-50). Downs with distal chromosome 21 trisomy do not get early onset disease.

Question 10

What is the strongest genetic risk factor for Alzheimer’s?

Answer 10

an allele of apolipoprotein E, e4 (ApoE4). It increases the risk 3x in heterozygotes and 15x in homozygotes. ApoE transports cholesterol to neurons. Most likely reason for association is that ApoE helps to remove or break down ABeta, and E4 is the worst at this. So genetics strongly supports the amyloid cascade hypothesis! – ie suggests Alzheimer’s is an amyloid disease: ABeta -> plaques -> death of neurons.

Question 11

What do amyloid plagues do in Alzheimer’s?

Answer 11

Plaques seem to cause local inflammation, ie they are treated by the body as ‘foreign’. Therefore at least to some extent, plaques cause the body to attack its own neurons and kill them. They also stimulate production of Reactive Oxygen Species (ROS) leading to local inflammation and tissue damage, vascular degeneration etc.

Question 12

What are ABeta oligomers?

Answer 12

These are when the monomers form balls rather than long fibres. All messier than fibrils because they don’t have well-defined shape or size or appearance. Also not clear if these are ‘on pathway’ to fibrils or ‘off pathway’.

Question 13

What do some experiments show with the effect of oligomers on Alzheimer?

Answer 13

There are lots of experiments which show that oligomers are more toxic than more mature fibrils or protofibrils.
This is the current most popular theory for Alzheimer’s: that the toxic species is the oligomers. This could be just because oligomers are smaller and therefore have more ‘ends’ to build on.

Question 14

What is the role of tau in Alzheimer’s?

Answer 14

Alzheimer’s is characterised by Aβ amyloid plaques and neurofibrillary tangles made of hyperphosphorylated tau.
Tau regulates microtubule assembly and disassembly. Therefore if you mess up tau, you mess up microtubules, and therefore transport along the axon. Transport along the axon absolutely needs microtubules!
Tau phosphorylation is governed by a balance between kinases and phosphatases. It may be that Aβ alters the activities of these enzymes, in particular the phosphatases. (Fairly) recent results suggest that Aβ can interact with certain receptors on neuronal membrane and produce intracellular signals – maybe this is how Aβ affects tau, and how Aβ is toxic.

Question 15

Why is making drugs for Alzheimer’s hard?

Answer 15

Getting volunteers for trials is hard! Drugs intended to improve symptoms: run 6-12 months (longer if looking for stabilisation). Drugs intended to produce cure: 18-24 months. Measure ‘severity’, cognitive performance (standardised scores). Needs a ‘significant improvement’ – sensitive measure and/or lots of patients. More recently also tested biomarkers e.g. ABeta or tau in cerebrospinal fluid; brain imaging for amyloid or brain volume. Might allow earlier detection.

Question 16

What do current Alzheimer’s drugs affect?

Answer 16

Neurotransmitters. Alzheimer’s causes death of neurons and loss of connections. Therefore you can slow down the effects of AD by strengthening signalling across the synapse, especially cholinergic. Use of an inhibitor of cholinesterase (cleaves acetyl choline) – makes signal last longer. Obviously this just affects the symptoms not the cause and cannot be a cure – it will only slow it down, maybe for 6-12 months. AD patients also often have antipsychotics or anti-depressants to manage behaviour. 4 cholinesterase inhibitors on market, plus one antagonist to NMDA glutamatergic receptor.

Question 17

What are the four possible targets for Alzheimer’s drugs?

Answer 17

Target 1: start of amyloid cascade.
Target 2: end of amyloid cascade.
Target 3: amyloid cascade (secretase Gamma)
Target 4: Tau

Question 18

How can Alzheimer’s drugs target the start of amyloid cascade?

Answer 18

How about the secretases? Good targets, because we know a lot about inhibiting proteases. a-secretase: no genetic link to AD. You need to increase activity, not decrease – some drugs in trial including epigallocatechin gallate (EGCG). b-secretase has nothing yet. y-secretase is the best hope, except that remarkably, g-secretase also cleaves the signalling receptor Notch, involved in neural development, heart development, and maturation of cells in various tissues. So a g-secretase inhibitor would have effects all over the body (and on the developing embryo) unless you could deliver it specifically to the brain.
Attempts to target not the g-secretase active site itself (2a), but its docking site for APP (2b) or a regulatory site (2c). Or try to influence where it cuts at the C-terminal end (2d/2e).
Surprisingly, several non-steroidal anti-inflammatory drugs make g-secretase change its preference, so that it makes more Aβ(1-40) and less of the more toxic Aβ(1-42). ie not inhibitors but modulators.

Question 19

How can Alzheimer’s drugs target the end of the amyloid cascade?

Answer 19

Can we do something to get rid of Aβ aggregates? a) Immunise using Aβ(1-42) and get the immune system to do the work. Great idea – does seem to reduce amyloid plaques. But 6% of patients developed meningoencephalitis. This is an example of active immunisation.
There are several antibodies against amyloid-beta:
1. Adcanumab that specifically recognises amyloid aggregates of Aβ, and not Aβ monomers. Sounds ideal. Several competitors.
July 2015: most of these trials were halted because of no effect or harmful effect. But two suggested that antibodies can slow down neurological decline. November 2016, Eli Lilley halted a trial in Phase III because of lack of effect; Pfizer halted another in 2017. March 2019: Biogen withdrew adcanumab for lack of effect. But they brought it back in Autumn 2019 because high doses looked like they might work. It was approved by the FDA in the US in June 2021 – three advisers to the FDA resigned because they felt this was a bad decision (and possibly politically motivated). It was withdrawn in January 2024 “for financial reasons”.
2. Lecanemab was approved by the FDA. A patients group apposed the approval, on the grounds that it had little benefit and was not safe. Neither has been approved in Europe or the UK for lack of clear efficacy.
3. Donanemab has also been announced by Eli Lilley (July 2023) that does the same thing. All of these antibodies reduce the amount of ABeta plaques and slow down progression of AD – but so far they do not appear to improve it.
These are examples of passive immunisation. These antibodies do seem to reduce the amount of ABeta plaques, and this does seem to slow down progression of dementia. So at least it suggests that this approach is feasible.

Question 20

How can Alzheimer’s drugs target the amyloid cascade?

Answer 20

(link to ApoE4). Reduce cholesterol – statins? Ongoing trials

Question 21

How can Alzheimer’s drugs target Tau?

Answer 21

Inhibit enzymes involved in tau phosphorylation; inhibit tau aggregation; stabilise microtubules. Ongoing trials (major trial halted in 2016 in Phase III due to lack of effect). 2019 shown that ABeta may bind to SirB receptor in mice. This could be the mechanism by which ABeta stimulates tau phosphorylation.
A surprising number of these drug trials actually made the symptoms worse. Some recent studies propose that Alzheimer’s is a type III diabetes /insulin dependence disease. It is certainly true that type 2 diabetes produces inflammation and vascular damage in the brain. This can produce a disease called vascular dementia, which may (or may not) be related to Alzheimer’s.
Or that Alzheimer’s is a consequence of mitochondrial dysfunction, and therefore therapies could target mitochondria.

Question 22

Why is ApoE4 a risk factor for Alzheimer’s?

Answer 22

Nobody knows. There are three main alleles of ApoE – 2, 3, 4. 4 increases the risk of Alzheimer’s and 2 protects against it. There are two main theories:
1. Gain of toxic function: ApoE4 is doing something bad (for example, making Beta-secretase cut APP to make ABeta(1-42) ).
2. Loss of useful function – maybe ApoE helps to clear excess ABeta, and ApoE4 is less good at this.

Question 23

How does Parkinson differ from Alzheimer’s?

Answer 23

Different from Alzheimer’s which mainly affects the neurotransmitter acetylcholine. Parkinson’s affects dopamine. The amyloid protein is called -synuclein, and it collects inside the neurons in ‘Lewy bodies’.

Question 24

What are the genetic causes of Parkinson’s?

Answer 24

• Gene for -synuclein. Function not clear – organisation of presynaptic vesicles? Common protein – up to 1% of all neuronal protein.
• Gene for leucine-rich repeat kinase LRRK2 (aka dardarin or PARK8). Function unclear but something to do with maintenance of dendrite structure (neurite outgrowth). No clear connection to Parkinson’s.
• Several others, but no obvious clues except for inflammation.
• In 2016, reported that TMEM230 is linked to Parkinson’s. Encodes a transmembrane protein of secretory/recycling vesicles, including synaptic vesicles in neurons – ie a defect in synaptic vesicles.

Question 25

What else can cause Parkinson’s other than genetic basis?

Answer 25

Also seems to be a link between gut microbiome and Parkinson’s – implies some kind of signalling between gut and brain.
Parkinson’s can be triggered by environmental factors (pesticides, head injury, metals) but evidence is not convincing. Some chemicals seem to protect, eg coffee, NSAIDs and smoking (!). Epigenetic link with DNA methylation. Some connection to mitochondrial damage? Probably has multiple causes.

Question 26

How is alpha synuclein ordered in the body?

Answer 26

In vivo alpha-synuclein is intrinsically disordered. Intrinsically disordered proteins are remarkably common. Roughly 1/3 of protein sequence in human genome; roughly 25% of human genes are predicted to be entirely disordered! – Though many of these become ordered on binding to partners. Intrinsically disordered proteins are more likely to form amyloid because they do not need to unfold first – lower energy barrier to forming amyloid.

Question 27

What is Kuru?

Answer 27

Very rare disease – limited to small part of Papua New Guinea and due to ritual cannibalism.
Ritual: women and children up to 8 ate brains and other organs of dead relatives. (Men ate other parts eg heart, so unaffected.) From about 1900 onwards, spread of kuru, mainly in women. Peak annual mortality of 2%, so very strong selection – some villages were left with almost no young women. If there is amyloid disease in the brain and brain is eaten, there is a risk of the disease being passed on, which is what happened.

Question 28

What is a selective sweep?

Answer 28

Selective sweep: strong evolutionary selection pressure leads to certain alleles becoming highly enriched. Some examples:
1. Lactose intolerance. As babies we have a gene for lactase, to digest milk (splits lactose into glucose and galactose). This gene is then turned off after weaning. The majority of the world’s population is lactose intolerant ie they have difficulty (as adults) digesting lactose, leading to stomach cramps, diarrhoea, and sickness. (However, they can eat cheese and yoghurt because much of the milk is already broken down.)
However, lactase gene is found in populations that herded cattle, and therefore had milk in adult diet – ie Europeans. Rare in Asian populations. Thought to have been lost in all humans and then re-acquired in dairy farming populations.
2. Host-pathogen interactions. Host and pathogen are in evolutionary battle, and influence each other.
Toxoplasma gondii is a common human parasite (and in other warm-blooded animals). Genetic analysis shows that a single clonal strain emerged in Europe and N. America within the last 10,000 years and is now spreading into S. America. Similar mechanisms probably influenced human genome too. eg Black death killed off nearly 1/3 of European population between 1347 and 1351, and probably produced a selective sweep of immune system Toll-like receptors (attachment site for Yersinia pestis).
3. Kuru

Question 29

What did Kuru do to the population genetics?

Answer 29

In kuru, the selective pressure led to almost complete change of G127 to V in local populations.
Highest frequency is found in elderly women. The remarkable thing about this is that the first recorded case of kuru was early 20th century, and cannibalism pretty much stopped by the 1960s or 1970s. So this genetic change across the population established itself in about 60 years or 3 generations.