Our First Working Hypothesis

We’re doing research differently. We use decentralized self-experimentation (guided by our own medical providers) and then pooling our data as a community. Traditional medical research tests one intervention at a time and one hypothesis at a time. But what if MECFS and Long COVID are systems problems and one intervention isn’t enough to tip the system into a new state? What if you start off with multiple hypotheses and then get clues for future study design to narrow down variables?

Why are we doing this way? - The story of our original hypotheses

There are stories of complete and sudden remission of symptoms of MECFS and Long COVID. Sometimes the person doesn’t know what caused the remission, while others have something to point to. Sometimes the remission is long term, and other times symptoms come right back. Understanding what is happening during these remissions could shed light on underlying pathophysiology of these conditions, but how can you study something that is unpredictable? 

We personally experienced and then collected anecdotes that antibiotics were a common cause of these “remission events”. We wondered, what if we could purposely trigger these remission events and be ready to collect at-home samples before, during, and after them to see what biomarkers change during the events?

We wondered, how could antibiotics cause remission and how could we make it more likely for an event to occur? We thought about different ways that antibiotics could affect the microbiome, immune system, etc. Although we collected ~50 anecdotes, it still seemed that antibiotic-triggered remission was rare, so we thought that we should be ready to measure as many biomarkers as possible, if an event happened. Most research experiments test one hypothesis and one intervention, but we imagined that if we were to try to do that with something that might be a rare event we might never actually be able to trigger it or capture a measurement of the right biomarker. 

We decided to take a different approach. We thought, let’s first prove that these remission events could be purposely triggered and get clues with our testing to drive future research. Our goal with our first experiment in March, 2023 was to make a remission event the most likely to occur and measure as many biomarkers as we could. In order to do this, we thought through hypotheses of what could be happening.

The original hypothesis diagram

‍

‍Many of the symptoms of ME/CFS are similar to those caused by evolutionary adaptive sickness behavior/response. Sickness behavior/response is the response that the body has when ill to try to get the person (or animal) to rest so that it can put more energy towards the immune system [Danzer 2009]. 

Sickness behavior has been connected to neuroglia and neuroinflammation [Wynne 2009]. Neuroinflammation has been found in ME/CFS and long COVID using PET scans with specific ligands [Nakatomi 2018] and neuroinflammation is caused by neuroglia inflammatory state [Bernaus 2020]. Renz-Polster [2022] has a good review of neuroglia and ME/CFS. 

When originally thinking about how antibiotics could trigger remission of MECFS, we broke it into four possibilities, illustrated in our hypotheses diagram and explained below. 

They kill “bad bacteria”.
They opened up a niche for “good bacteria”
They have direct effects in the body.
They have direct effects on the brain. 

Examples of changes due to modifying bad bacteria include lowering lactate producers [Abedi 2020], GABA producers [Strandwitz 2019], bacteria with bacteriophages in them [Sutton 2019], and/or gram negative bacteria that can lead to lipopolysaccharide (LPS) leaking into the bloodstream [Mohr 2022]. Some possibilities for bacterial products that could come from opening a niche and positively affect the body include increasing tryptophan metabolite producers [Dehhaghi 2019], butyrate or propionate producers [Silva 2020], serotonin producers [Yano 2016], dopamine producers [Ekker 2022], and/or BH4 producers [Belik 2017]. 

If remission events are caused by killing bad bacteria, it’s possible that cell danger signals decreased, turning off systemic inflammation, leading to lower neuroinflammation/neuroglia activation [Tate 2022]. Similarly, by increasing good bacteria, the products that they release into the bloodstream could interact with the immune system and decrease inflammation and neuroinflammation. 

Another set of hypotheses is that the bacterial change or direct effects on the brain from the antibiotics could have caused a sudden swing in neurotransmitters, such as dopamine, GABA, or serotonin. Dopamine in particular could account for the change in our perception of colors [Kuhn 1995]. For GABA, Amoxicillin is known to have direct effects on GABA receptors in the brain [Skelley 2019]. Since GABA is an inhibitory neurotransmitter, if it was decreased it could lead to increased neuronal hyperexcitability.

Other direct effects in the body or brain could be due to killing persistent infections (bacteria directly in the body or brain, rather than the gut.) This could lead along the same path as the first sets of hypotheses (lowering inflammation, neuroinflammation, and sickness behavior/response.)

Patient experience leads to new hypotheses
‍
Self-experimentation can lead to unique insight. One of the first Remission Biome self-experimenters, Kat Boniface, a PhD student at UC Riverside, was meticulous with her observations while taking Amoxiclav. She noticed how suddenly symptoms could drop off and paid attention to the timing of how fast it happened after starting the Amoxiclav. That led her down a research rabbit hole, thinking about how the experience could possibly be related to modifying glutamate in the brain.

It turns out that the Clavulanic Acid portion of Amoxiclav can directly modify glutamate systems in the brain! High extracellular glutamate levels have been found in MECFS. 
Independently, another of our team members, Nick Melia, also came across the glutamate research. Kat and Nick are now working together to put the pieces together for this new hypothesis.