I’m sure you’ve heard of Alzheimer’s Disease (AD). It’s a neurodegenerative disorder most commonly characterized by dementia, loss of judgement, and overall, the loss of cognitive function. It affects over 5.8 million Americans or 11% of all Americans over the age of 65. And in 2018 alone, Alzheimer’s accounted for 122, 019 deaths, making it the sixth leading cause of death in America and the fifth leading cause of death among Americans above 65.
So, as you might guess, lots and lots of research has been done to find treatments and cures for the disease, ranging from small drug molecules that target cells in the brain to not so successful genome engineering approaches.
“The reality of this disease is that there is no remission, there’s no stopping it, there’s no slowing it down, there’s no cure, and there are no survivors”
- Suzanne Tackett, Alzheimer’s patient
But before we go into that, let’s discuss a bit of pathology and genetics.
Why Does Alzheimer’s Develop?
The truth? We still don’t properly know for sure why Alzheimer’s develops.
But… we do have a really good guess at it, through a theory known as the Amyloid Cascade Hypothesis.
Amyloid Cascade Hypothesis
Essentially, researchers have observed surprisingly high levels of a protein known as beta-amyloid (Aβ), or amyloid-beta peptides, in the brains of patients with Alzheimer’s. Though normally associated with neuroplasticity, this beta-amyloid protein accumulates in the brain as oligomers and amyloid plaques, which all injure neuronal synapses. And this is what leads to the massive amounts of neuronal damage that cause the symptoms seen in Alzheimer’s.
But why does the protein accumulate in the first place?
It has to do with mutations in a few specific genes.
The first of these is the amyloid precursor protein (APP) gene. This gene encodes (as the name suggests) the amyloid precursor protein, which is cleaved to form beta-amyloid. In the brain, APP is thought to assist with the processes that direct neuron movement in the early stages of development. When this protein is cleaved by enzymes, many smaller fragments are creating, including amyloid-beta peptides. However, when APP is cleaved incorrectly, a variant of beta-amyloid known as amyloid-beta 42 (Aβ₄₂), which is found to be “stickier”, or more likely to accumulate, and thus contributes to the Aβ deposits. Because of the huge impact the APP protein has on the production of Aβ, it’s not surprising that mutations in the APP gene can cause Aβ₄₂ production. However, mutations in APP are only responsible for less than 10% of all early-onset AD cases.
Another gene associated with early-onset AD is the presenilin-1 (PSEN1) gene, which codes for the presenilin-1 protein that acts as a cleaving subunit on the gamma-secretase (γ-secretase) complex, which is responsible for cleaving other proteins into smaller peptides through a process known as proteolysis. The γ-secretase complex is known best for its role in cleaving APP into smaller peptides, including Aβ. So, mutations in PSEN1 are the most common cause of early-onset Alzheimer’s disease, and accounts for over 70% of all cases! This is because errors in the protein sequence of PSEN1 can cause improper processing of APP, leading to the overproduction of Aβ₄₂.
The last main gene associated with early-onset AD is the presenilin 2 (PSEN2) gene. Presenilin 2 combines with other enzymes that help to process proteins that transmit chemical signals from the cellular membrane to the nucleus, including the amyloid precursor protein. However, mutations in PSEN2 are responsible for less than 5 percent of all cases.
Any of these mutations can contribute to the production and accumulation of Aβ₄₂, which causes synaptic and neuronal damage that eventually leads to dementia and other symptoms associated with Alzheimer’s Disease.
So, if we know all these genes and the impact they have, why can’t we just treat AD?
Well, because these genes have only been linked to early-onset Alzheimer’s disease (symptoms begin developing before 65), which only makes up 5–10% of all Alzheimer’s disease cases.
However, the majority of cases fall under sporadic, or late-onset, Alzheimer’s Disease, which occurs in patients 65+.
Sporadic Alzheimer’s Disease
Sporadic AD is a whole other story.
The problem is, there has been no clear genetic link found for patients suffering from this type of Alzheimer’s. But… that doesn’t mean there aren’t any candidates.
In fact, the most likely genetic factor associated with late-onset AD (LOAD)is a gene known as apolipoprotein E (APOE). It codes for (you guessed it!) apolipoprotein E, a major cholesterol carrier that supports lipid transport and injury repair in the brain. The APOE gene has three main alleles (versions): E2, E3, and E4. Out of these, the E4 allele is the strongest risk factor for both LOAD and (surprisingly) early-onset AD. In fact, the frequency of AD patients is 91% for those that carry two copies of the E4 allele, 47% for those that carry just one copy, and only 20% for non-carriers, proving that E4 dramatically increases the risk of developing AD.
Now, even though E4 is the strongest risk factor for early-onset AD out of all three alleles, in the grand scheme of things, E4 doesn’t play as large a role in the risk associated with early-onset AD compared to the other genetic factors.
So… once again… even if we were to use genetic therapy on the APOE gene, it would only reduce the risk of developing LOAD but would have barely any impact on early-onset patients. Furthermore, it is important to note that E4 only increases the risk of developing AD, but does not guarantee it. So, even after editing this gene, it is not guaranteed that we can prevent LOAD.
So… we need to find another genetic link that has effects on both LOAD and early-onset AD, to essentially develop a generalized Alzheimer’s disease treatment for all patients.
That’s where another biomarker of Alzheimer’s Disease comes in.
So, your brain has special immune cells of its own, known as microglia. Normally, their function is to clear debris from the brain, as well as protecting the brain from invading pathogens and killing off damaged cells to prevent cancer.
But in the case of Alzheimer’s, the microglia often go overboard in their role. This happens because they become overactivated by the same amyloid-beta peptide deposits, sending the microglia into a neuron-killing and damaging frenzy.
And all of that happens because of dysfunctional receptors on the surface of microglia known as TREM2.
TREM2 belongs to a family of receptors known as triggering receptors expressed on myeloid cells (TREM), and codes for a 230 amino acid glycoprotein. The gene is expressed in a subgroup of myeloid cells including dendritic cells, granulocytes, and macrophages, while it is exclusively expressed in microglia in the brain. The receptor acts mainly through the intracellular adapter DAP12 and is incapable of initiating intracellular signalling without it.
The receptor recognizes a variety of ligands (or molecules it binds to), including lipopolysaccharides, lipoproteins, and many versions of apolipoprotein, including APOE and clusterin (both of which are genetic risk factors for LOAD). And… beta-amyloid has also been characterized as a ligand for TREM2, capable of binding directly to the receptor & activating TREM2 signalling.
Upon recognition of any of these ligands, TREM2 carries out several functions, including causing an increase in phagocytosis (basically eating up damaged cells and invaders). In vitro experiments found that inhibition of TREM2 in microglia resulted in decreased phagocytosis of apoptotic neurons (essentially damaged/dead neurons left behind), cellular debris, including Aβ, and bacterial products, while increased expression raised phagocytotic rates.
TREM2 was also found to modulate inflammatory signalling in microglia, where knockdown of TREM2 signalling increased TNFα cytokine and NO synthetase-2 transcription and mediated the switch from a homeostatic (maintaining balance) to neurodegenerative microglia phenotype. This, along with abnormal phagocytosis, most likely contributes to the neuroinflammation and microglial activation observed in TREM2 mutations.
TREM2 Mutations and Effects
By using methods like whole-genome sequencing and genome-wide association, researchers have found several rare TREM2 variants that were associated with increased risk of AD. The most commonly studied of these is the R47H variant, which expresses a single-nucleotide polymorphism (SNP; change in just one letter) causing a change from arginine to histidine (both are amino acids) in amino acid #47. This mutation alone triples the risk for developing LOAD, making it comparable in effect to the E4 mutation in APOE. Many other variants were also identified, including R62H, D87N, T96K, L211P, and R136Q.
Most studies determined that these SNPs didn’t effect the structure of the TREM2 receptors themselves, but instead altered their affinity for its ligands.
The absence of functional TREM2 caused by these SNPs was observed to cause an increase in amyloid seeding and decreased microglial clustering around newly seeded plaques. This suggests that early amyloidogenesis is accelerated because the phagocytotic clearance of amyloid deposits is compromised as a result of dysfunctional TREM2 variants. This lack of microglial clustering also led to defects in plaque compaction, proliferation of microglia, and an increase in the levels of dysfunctional neurons.
On the other hand, it was demonstrated that levels of both soluble and insoluble Aβ₄₂ were significantly decreased in mice that expressed functional human TREM2 transcripts.
And so, all of this proves the large impact mutated TREM2 can have on the progression of Alzheimer’s Disease. And so, it can also be hypothesized that treating these mutations could allow for prevention, or at least reduction in the risk, of Alzheimer’s Disease.
Now that we’ve established that we might be able to treat Alzheimer’s disease by correcting the TREM2 gene, how exactly do we do that?
Well, for the editing part itself, we use an upcoming gene editing method known as base editing.
So I’m sure you’ve heard of CRISPR. It’s been gaining a lot of hype in the biotech industry for over a decade as the future in genome engineering and treating genetic diseases.
Well, though CRISPR can be described with all of those, the system itself still has its flaws. One of those is the fact that the editing itself isn’t always completely accurate, especially since it can be a bit overboard for fixing single nucleotide polymorphisms.
To solve this problem, some researchers are experimenting with base editing — a gene-editing approach that uses some of the components from CRISPR with other enzymes to treat point mutations, or SNPs, in using DNA templates. DNA base editors comprise fusions between the Cas nuclease (responsible for cutting the DNA strands) and a base modification enzyme that only works on one strand of DNA, unlike normal CRISPR systems that work on both strands.
After binding to the target location on the DNA, base pairing between guide RNA (identifies the editing site) and the DNA leads to a small displaced segment of single stranded DNA looped up, where DNA bases are modified by a deaminase enzyme.
Using this approach, we can make accurate edits of single letters at a time, allowing us to accurately correct single nucleotide polymorphisms, including the ones responsible for dysfunctional TREM2 variants.
So now that we’ve decided how to edit the genome, how do we get this editing machinery into the microglia in the first place?
Adeno-Associated Virus Vectors
Adeno-associated viruses (AAVs) are a class of viruses regularly used as vectors for different gene editing experiments. The viruses are highly stable and rarely induce an immune response, making them the perfect candidate to target cells without placing the patient in danger.
In this case, we could use a variant of AAVs called AAV6, which has been designed specifically to target microglial cells.
So, what do I expect to happen if the experiment is successful?
Well, quite simply, since microglia would now be able to clear beta-amyloid effectively, we expect less amyloid-beta deposits. Furthermore, since the microglia would no longer be in a neurodegenerative phenotype, we would also observe less inflammation. And lastly, we would also expect the cells to be able to properly destroy dysfunctional neurons.
And, we can hypothesize that all of these effects together would lead to an overall prevention, or at least slowing down, of the progression of Alzheimer’s disease in patients with TREM2 mutations.
- Alzheimer’s disease is associated strongly with the buildup of beta-amyloid peptides
- In early-onset Alzheimer’s, the accumulation is caused by mutations in either of the APP gene, the PSEN1 gene, or the PSEN2 gene
- In late-onset Alzheimer’s, no clear genetic link has been found, but an association has been discovered with the APOE gene
- Out of all the variants of APOE, E4 greatly increases the risk for developing Alzheimer’s disease
- In Alzheimer’s disease, microglia, the immune cells of the brain, stop clearing beta-amyloid peptides and start releasing excessive amounts of cytokines, causing neuroinflammation
- The microglial dysfunction is caused by mutations in the TREM2 gene, which codes for a receptor on microglia
- By treating the TREM2 mutations with base-editing, which is similar to CRISPR, we can treat Alzheimer’s disease
Check out my video explaining this proposal:
On a More Personal Note
I am a 17-year old currently obsessed with the science behind aging, and if we could live forever (because, let’s be honest, no one wants to die), along with how we can treat neurological disorders like Alzheimer’s disease.
If you want to read more of my articles, follow me here on Medium and check out some of the other articles I have written, on various topics from AI, to gene editing, to life and philosophy. To find out more about me, check out my website, and my Twitter.
You can check out my draft research proposal and short video explaining the proposal here.
That’s all for me for now. See you next time!