Hacker News Reader

Comment by aaavl2821 | Hacker News Reader

Very informative and thought provoking post

If I understand correctly, APOE helps viruses enter neurons. How does APOE4 behave differently than other isoforms in this regard? APOE itself isn't implicated in AD risk, just the APOE4 isoform. And I couldn't tell from your post or a brief lit search, what's the current thinking on the relationship btw APOE and amyloid beta?

I think APOE4 is associated with higher levels of beta amyloid but I don't know why. You mention a study showing beta amyloid traps herpes simplex? Is there a mechanism by which APOE4 drives more beta amyloid aggregation to fight viral infections, but as an unintended result leads to AD?

I've briefly followed some of the recent papers on a viral link to AD but haven't spent much time on it, your post has inspired me to do some more research!

View on HN

Here are a few major points :

* APO-E isoform, that predicts alzheimers risk, correlates with how much beta amyloid is present in the brain

* It also correlates with how much Herpes Virus is present : https://www.theatlantic.com/science/archive/2018/07/herpes-v... "Readhead and his colleagues have shown that the more E4 copies someone has, the more HHV–6A and HHV–6B viruses they are likely to have in their brains."

* We also know that if you knock APOE out in a mouse, and then infect it, there is WAY less HSV penetration in the brain (see figure 1, and note that the Y-axis is logarithmic!) : http://jvi.asm.org/content/76/23/12394.full

* Also, APOE isoforms predict whether or not Hepatitis B virus will give you liver cancer. Note that its the E3 genotype that is risky for this, yet its the E4 genotype thats bad for Alzheimers. I take this to be evidence that hepatitis viruses and herpes viruses likely have different binding affinities to these two isoforms. https://www.ncbi.nlm.nih.gov/pubmed/26823800

* Someone should do some drug screening to eliminate binding between known viral capsids and APOE variants. cough YC Bio

Seems like there is still a lot to learn about the underlying biology, though very provocative findings thus far. I may be misunderstanding but I'd think that if beta amyloid traps herpes, then higher levels of beta amyloid would correlate with lower levels of virus, but it seems that isn't the case?

Would antiviral therapy be more effective that blocking capsid-APOE interaction? It's my understanding that blocking protein protein interactions is challenging, although inhibiting that interaction would be a nice specific way to block the mechanism

It's a fair point, for sure. I'm not so sure how antivirals work but that is probably a more direct pathway to target.

Alternatively, maybe we could train our immune systems to detect when a lipoprotein is transporting malicious cargo. Ideally, if we could realiably distinguish between healthy LDL particles and lipoviral particles, maybe we can find surface epitopes!

I wonder if the reason foam cells are forming in atherosclerosis is because the immune system suspects that they may be infected and carrying payloads.

Certainly seems like an interesting avenue, though I don't know enough to speculate as to whether there are surface epitopes (i dont even know how / where on the LDL the virus binds haha)

The relationship btw HSV, APOE, amyloid beta and AD is certainly intriguing but it seems quite messy at this point. Would imagine there's a lot of work to do to unravel the mechanism before we get to the point where we can think about how best to drug whatever target emerges. Unfortunately mouse models in AD are terrible, so studying the biology may be very tough, although it seems iPSC derived neurons are proving to be decent models.

One thing I like about a lot of HN discussions, as opposed to those on many other sites on the web, is that the people involved have a clear sense of what they do and don't know.

> the people involved have a clear sense of what they do and don't know.

As someone who has, on more than one occasion, called out top-posts that were flat-out wrong, but conformed to HN's biases (I was 100% certain of this because the posts concerned either my area of expertise or my first-hand experiences) - I will disagree with you. The occasions made me wonder about the other HN posts where I'm not an expert.

HN isn't anywhere near as free of the usual human love of gossip, valuing social pecking order over objective facts, etc as it imagines itself to be.

But it's a little like that saying about democracy being "the worst thing ever -- except for everything else". In spite of HNs serious delusions about how objective etc it is, it still remains vastly more objective, rational etc than anything else I've found.

Yes, that's the point I was trying to make.

your question: higher levels of beta amyloid would correlate with lower levels of virus. A metaphor for answering it: if the protection arrives late and the harm is done then the number of protectors can't be used to predict the harm. So to anaylize the correlation one has to know if the effect of the protection happen before it makes the harm or not. That is the speed here is an essential ingredient. Emperator caveat: I applied abstract reasoning and a metaphor, don't know about the specific details of how beta amyloid is generated.

Perhaps the presence of the virus causes the beta amyloid to be produced, or to not be cleared (because it's being used to trap the virus)? I'm no biologist but spitballing on why you might see a direct relationship vs an inverse one.

Reply on news.ycombinator.com