randombio.com | commentary
Wednesday, October 16, 2019

Beta-amyloid is good for you

Does a common childhood virus lead to Alzheimer's disease later in life?

S ome years ago I did a study to find out what proteins could bind to beta-amyloid, the protein that has come to be synonymous with Alzheimer's disease, so we could figure out its normal function. I didn't find much: an adenosine receptor and a signaling protein called 14-3-3. My boss told me to change the paper to portray my result as potential treatment so he could patent it, which derailed the entire project, and I abandoned it. But I wonder now what would have happened if a virus protein had shown up. Would I have just thought it was a mistake? Or would I have followed through with it and gotten fired?

Those are the choices you have to make in science. If you don't have tenure, you have to do what the pencil-pushers and PHBs want. But now, in the past two years, researchers in three major studies have discovered that beta-amyloid, the little protein that everybody thought caused dementia, and which drug companies have spent billions trying to get rid of in Alzheimer patients, is actually good for you.

The Taiwan Study

The first study was done by Nian-Sheng Tzeng et al.[1] who looked at health records of 33,448 people in Taiwan, 8362 of whom were diagnosed with an active herpes simplex virus (HSV) infection. They found that those patients were 2.564× as likely to get dementia, and giving them an antiviral agent reduced the incidence of dementia within ten years from 28% to 5.8%.

 Group        Number of patients        With dementia        Hazard ratio        p value      
HSV 8362 Not stated 2.5640.001
Control 25086Not stated
HSV+AV 7215 419 (5.8%) 0.0920.001
HSV−AV 1147 325 (28.3%)

If viewing on a cell phone, drag table left or right to scroll.

Now, there are many things wrong with this study. They didn't distinguish between HSV1 and HSV2. They didn't separate Alzheimer's from vascular dementia or other types of dementia. And the same author was known for writing a number of other papers claiming a 4.2× increased risk of dementia from burning charcoal, a 3.4× increased risk from having ADHD, and a 2.54× increased risk from periodontitis. But if true, it was a remarkable result.

The Bioinformatics Study

Then Ben Readhead et al.[2] did a DNA network analysis of the Alzheimer-associated virome, which is to say, all the viruses in the brain regardless of whether they're active or not, in four brain regions known to be affected early in Alzheimer's. They also collected clinical dementia rating scores, amyloid plaque densities, and Braak scores, which are measures of the severity of AD done at autopsy.

The sequences included RNA coding for proteins that were expressed in neurons. This was done to measure neuron loss. They also looked at apoptosis genes to examine programmed cell death, PARP1 expression to measure DNA repair, and many other genes thought to be important in AD. From these numbers they constructed “probabilistic causal networks,” which is a fancy term for calculating many, many Pearson correlation coefficients between everything and everything else.

The results implicated three viruses: HHV-6A, HHV-6B, and HHV-7. These are human herpesviruses that are between 95 and 100% seroprevalent, which means almost everybody has antibodies against them. Virologists call them Roseoloviruses. They cause a rash called roseola infantum which happens in about 30% of children. HHV-6B is also associated with mesial temporal lobe epilepsy because it gets reactivated in the CNS later in life.[4] Seizures in this brain region are a prominent feature of early stages of AD.[5]

So the question becomes: does HHV cause AD, or is it an opportunistic infection? Maybe, you might say, the blood-brain barrier gets more permeable in AD patients, as is known to happen, and that allows the viruses to enter, and the brain gets infected. Or maybe it's like shingles, where the chickenpox virus lies dormant for decades and then starts attacking neurons later in life.

If this theory is true, antivirals might be effective treatments for AD. Childhood vaccines against HHV-6 and 7 would be even better. But there's a possibility that almost anything that activates the innate immune response, including viruses, toxins, or bacteria, could trigger out-of-control neuroinflammation and lead to AD.

The Eimer Study

Just by wild coincidence, Eimer et al.[3] happened to be studying how beta-amyloid oligomers bind to herpesvirus (HSV1, HHV-6A, and HHV-6B), and wrote a paper in exactly the same journal the same week. These oligomers are made in the lab from synthetic beta-amyloid, which you can buy, and oligomers are thought to be the toxic form of it. Eimer et al. found that oligomers bind strongly to herpesvirus surface glycoproteins. They concluded that beta-amyloid is not toxic at all, but is an anti-microbial peptide made by the cell to protect against virus particles by entrapping them in amyloid plaques. Indeed, viruses are found in the plaques, and the group reported that beta-amyloid protects against encephalitis produced by HSV1 in mice. Which is to say, the AD mice exposed to HSV1 lost less weight and lived a lot longer than wild-type mice exposed to HSV1. AD mice living longer than controls never happens, and it's a remarkable finding.

Beta-amyloid also supposedly protects against influenza virus. It would be interesting to check if AD patients have a reduced incidence of the flu. You might not think influenza virus would be found in the brain, but in fact it's an important cause of acute encephalitis in children, and this encephalitis has a very high mortality rate.


They also said that beta-amyloid might be toxic too, and that active herpes infections also may be working by accelerating amyloid deposition that contributes to AD progression. The beta-amyloid-as-a-toxin theory doesn't contradict their main finding: it could do both.

Now, this was all done in 5XFAD transgenic mice. It's easy to find molecules that accelerate or inhibit amyloid deposition in transgenic mice, especially 5XFAD mice, which have (as the name suggests) five separate mutations that guarantee they'll have massive amounts of beta-amyloid. There are hundreds of papers about drugs that inhibit amyloid expression or toxicity in transgenic mice—so many that it's hard to publish them anymore.

The authors even admit that transgenic mice totally suck as models for AD, though they don't put it quite that way. Also, it's unlikely that beta-amyloid is specific for this particular virus. But I never saw a clearer demonstration that beta-amyloid is there for a reason.

The only way to be sure, short of injecting people with harmful viruses to see if they get Alzheimer's, would be to do a clinical trial and give AD patients antiviral medications. Those are happening now: two with valacyclovir in Phase 2, and one with efavirenz in Phase 1.

Back when I was first instructing my research assistants on how to handle samples from AD patients, I always made them wear full face shields, lab coat, and gloves. Some day, I told them, somebody will come along and prove that Alzheimer's disease is caused by a virus, and you'll thank me.

They laughed. They thought I was kidding.

There's a lesson in there somewhere. Anything scientists say that is bad could turn out to be good for you, and vice versa. If it's in a scientific journal, it's in flux. Take findings out of the research literature and use them to make public policy, or spend billions of dollars trying to use them, and the paradigm could shift under your feet. Fans of the climate change battle should take note.

1. Tzeng NS, Chung CH, Lin FH, Chiang CP, Yeh CB, Huang SY, Lu RB, Chang HA, Kao YC, Yeh HW, Chiang WS, Chou YC, Tsao CH, Wu YF, Chien WC. (2018). Anti-herpetic Medications and Reduced Risk of Dementia in Patients with Herpes Simplex Virus Infections-a Nationwide, Population-Based Cohort Study in Taiwan. Neurotherapeutics. 15(2), 417–429. doi: 10.1007/s13311-018-0611-x. PMID: 29488144 PMID: 29488144 PMCID: PMC5935641 Link

2. Readhead B, Haure-Mirande JV, Funk CC, Richards MA, Shannon P, Haroutunian V, Sano M, Liang WS, Beckmann ND, Price ND, Reiman EM, Schadt EE, Ehrlich ME, Gandy S, Dudley JT. (2018). Multiscale Analysis of Independent Alzheimer's Cohorts Finds Disruption of Molecular, Genetic, and Clinical Networks by Human Herpesvirus. Neuron 99(1), 64–82. doi: 10.1016/j.neuron.2018.05.023. Epub 2018 Jun 21. PMID: 29937276 Link

3. Eimer WA, Vijaya Kumar DK, Navalpur Shanmugam NK, Rodriguez AS, Mitchell T, Washicosky KJ, György B2, Breakefield XO, Tanzi RE, Moir RD. (2018). Alzheimer's Disease-Associated β-Amyloid Is Rapidly Seeded by Herpesviridae to Protect against Brain Infection. Neuron 99, 56–63. doi: 10.1016/j.neuron.2018.06.030. Link

4. Fotheringham J, Donati D, Akhyani N, Fogdell-Hahn A, Vortmeyer A, Heiss JD, Williams E, Weinstein S, Bruce DA, Gaillard WD, Sato S, Theodore WH, Jacobson S (2007). Association of human herpesvirus-6B with mesial temporal lobe epilepsy. PLoS Med. 4(5),e180. PMID: 17535102 PMCID: PMC1880851 DOI: 10.1371/journal.pmed.0040180 Link

5. Lam AD, Cole AJ, Cash SS. (2019). New Approaches to Studying Silent Mesial Temporal Lobe Seizures in Alzheimer's Disease. Front Neurol. 10, 959. doi: 10.3389/fneur.2019.00959. PMID: 31551916 PMCID: PMC6737997 Link

oct 16 2019, 4:55 am. updated oct 24 2019

Related Articles

The theory that just won't die
Beta-amyloid has been studied for 30 years, and we are no closer to a cure.

Mutant amyloid precursor proteins found in late-onset Alzheimer's disease
What did they find, and what does it actually mean?

Solanezumab drug trial failure: curtains for the beta-amyloid theory?
The solanezumab clinical trial failure hasn't eliminated beta amyloid as a potential cause of Alzheimer's disease.

On the Internet, no one can tell whether you're a dolphin or a porpoise

book reviews