Serotonin reduction in post-acute sequelae of viral infection, 2023, Wong, Cherry et al

Serotonin reduction in post-acute sequelae of viral infection

Highlights

• Long COVID is associated with reduced circulating serotonin levels
• Serotonin depletion is driven by viral RNA-induced type I interferons (IFNs)
• IFNs reduce serotonin through diminished tryptophan uptake and hypercoagulability
• Peripheral serotonin deficiency impairs cognition via reduced vagal signaling

Summary
Post-acute sequelae of COVID-19 (PASC, “Long COVID”) pose a significant global health challenge. The pathophysiology is unknown, and no effective treatments have been found to date. Several hypotheses have been formulated to explain the etiology of PASC, including viral persistence, chronic inflammation, hypercoagulability, and autonomic dysfunction.

Here, we propose a mechanism that links all four hypotheses in a single pathway and provides actionable insights for therapeutic interventions. We find that PASC are associated with serotonin reduction. Viral infection and type I interferon-driven inflammation reduce serotonin through three mechanisms: diminished intestinal absorption of the serotonin precursor tryptophan; platelet hyperactivation and thrombocytopenia, which impacts serotonin storage; and enhanced MAO-mediated serotonin turnover. Peripheral serotonin reduction, in turn, impedes the activity of the vagus nerve and thereby impairs hippocampal responses and memory.

These findings provide a possible explanation for neurocognitive symptoms associated with viral persistence in Long COVID, which may extend to other post-viral syndromes.

Graphical abstract


https://www.cell.com/cell/fulltext/S0092-8674(23)01034-6#

 

Quite a big study with a ton of authors, many of which are well known in the Long-Covid field. The study is already being covered by the big news outlets in America.

https://www.nytimes.com/2023/10/16/health/long-covid-serotonin.html
https://www.bloomberg.com/news/arti...t-long-covid-s-body-wide-symptoms-study-finds
https://www.statnews.com/2023/10/16/long-covid-brain-fog-serotonin-interferons/
https://www.expresshealthcaremgmt.c...ng-to-a-potential-cause-for-brain-fog/165031/
https://medicalxpress.com/news/2023-10-serotonin-reduction-covid-symptoms.html


A thread by one of the authors:
www.twitter.com/MaayanLevy_Lab/status/1713929997155811789

 

Interesting findings. They'll be testing SSRIs as well, but I'm not aware they've been of much use in ME. This paper says the following.

https://www.sciencedirect.com/science/article/abs/pii/S0149291819300712

 

Haven't read this paper yet, but looks important, so #1 for this morning.

Replicates findings in Integrative Plasma Metabolic and Lipidomic Modelling of SARS-CoV-2 Infection in Relation to Clinical Severity and Early Mortality Prediction (2023, International Journal of Molecular Sciences). They were able to predict mortality out to 61 days from blood test at 24 hours, with an AUC of 0.96 and similarly long covid.

On the subject of serotonin, that paper said —

 

Yeah I thought that too and I feel like we've seen some long covid metabolomics before and not seen serotonin (maybe that dutch paper about kynurenine)

Their starting point is a meta analysis of previous metabolomics papers for acute COVID-19 where they see reduced serotonin and use it as justification for the rest of the work in this paper :



Then see it again in a cohort of patients with acute viral infection generally - these patients had sepsis and were in intensive care. A similar looking effect size quite a night and day difference. They then go on do various viral challenges in mice and see a similar phenotype. When they use genetically altered mice deficient in various innate immune signaling (TLR3 / STAT1) genes the phenotype is rescued and serotonin levels go back up. They argue therefore that serotonin deficiency is a general feature of viral infection which is mediated by sensing of viral RNA and induction of type 1 inteferon.


There's a lot more to this paper and seems like an interesting one to read through. Obviously I would like to see this done in ME/CFS, I wonder if we'd see a phenotype quite as strong as this.

 

Did they consider estrogen , particularly in females
Serotonin and estrogen ' s relationship modulates many things .
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1327664/

 

Just a thought, why then did that DBRCT of ivermectin or metformin or fluvoxamine in LC show only metformin had a beneficial effect? Fluvoxamine an SSRI didn’t work

 

These were my exact first thoughts as well.

The second thing is that the idea behind the paper is somewhere along the lines of "what goes wrong in those with a very severe actue infection could be similar to what is happening in LC on a permanent basis". If that is the case there has to be an explanation for people seemingly fully recovering from acute Covid to then a couple of weeks later developing Long-Covid. SARS-COV-2 somehow going unnoticed and somehow establishing a reservoir during this time or the infection reactivating a latent virus which for whatever reason (possibly in combination with a reservoir of SARS-COV-2) staying somewhat permanently activated or permanently activating a pathway seem to be the most straightforward yet highly speculative answers to me.

One of the more surprising conclusions to me, given that this group is also very focused on viral persistence, is that they didn't care too much about the stool samples (p=0.06) and rather want to target mechanisms upstream from this with an SSRI. Perhaps they themselves didn't think the results were too remarkable. After all a little graph with a p-value only tells you very little. Was there perhaps a big decrease in the stool RNA load for patients that were further down the road? Or perhaps the focus of this paper should be a different one than what they are already doing elsewhere.

What I seriously do like about the study is the level of rigor they went through, possibly one of the most rigorous studies I've seen in Long-Covid. Of course it's still possible to find noisy phenomena on different levels.

Instead of guessing how these new Serotonin could be noise, given the rigor of the study, I will play devils advocate to see why other studies possibly didn't corroborate these findings. The most straightforward explanation I see, to argue why their results aren't noise, but still aren't reflected in some other studies is that some sort of viral persistence (SARS-COV-2, enterovirus gut reservoir or some other virus) is only happening in a subset of patients and you have to pick this correct subset.

UNCOVR study
-explained above by @Hutan

Immunometabolic rewiring in long COVID patients with chronic headache
-A study looking at post-Covid headache patients that found elevated levels of serotonin throughout various time points, from the acute phase till 3 months later. They collected samples a minimum of 8 days prior to post-Covid headache onset and then at another 4 time points with the latest time point being a minimum of 3 months post LC headache onset.
-They found that those which developed a headache had activated inflammatory paths including TLR (TLR-4/8/2/1/10) and IFN pathways (including elevated STAT1 levels). In particular IFNγ and IFNγ-associated downstream cytokines were detected in LC headache group throughout sampling.
-Perhaps it could be possible that whatever causes the dysregulation of tryptophan biosynthesis could somehow spin either ways or that different inflammatory pathways cause different phenomena with post-Covid headaches belonging to a different phenotype. Indeed this study found reduced levels of kynurenine (whilst those in the Cherry et al study normalised after the acute infection).

Fluvoxamine (mentioned by @leokitten above)
-Two studies by Bramante et al. One looking at viral load in acute Covid, the other looking at the risk of developing Long-Covid. As Cherry at al point out "the effectiveness of SSRIs in acute COVID-19 has been a subject of debate". However, anecdotal evidence (both in LC and ME/CFS) as well as smaller studies don't look convincing. Perhaps these mechansims are too far upstream to be targeted so naively or once again the "correct" phenotype has to be chosen?

Which other studies are there that contradict the above results and are at least decently conducted?

 

Moved post

Scientists Offer a New Explanation for Long Covid

A team of scientists is proposing a new explanation for some cases of long Covid, based on their findings that serotonin levels were lower in people with the complex condition.
In their study, published on Monday in the journal Cell, researchers at the University of Pennsylvania suggest that serotonin reduction is triggered by remnants of the virus lingering in the gut. Depleted serotonin could especially explain memory problems and some neurological and cognitive symptoms of long Covid, they say.

Gift Article Link
https://www.nytimes.com/2023/10/16/...U7JFK2pLqxOmm3-fiCJl0ENQ16esJw&smid=url-share

 

https://twitter.com/user/status/1714284227171389784



Disappointing it took more than a year to get it out! Submitted on August 2022!
cell.com/cell/fulltext/…

 

So it appears that most people (in general, in the media) are focusing on the serotonin and Prozac part, and very little on the platelets and interferon part, or the tryptophan.

Figures.

I don't know but if there's a deficit of some important molecule I wouldn't try some harebrained way of juicing up what little of it there is, one that we know doesn't work, and rather focus on stopping the deficit since it clearly has a cause. But maybe I'm just weird.

 

I have two questions for anyone who might know the answer or have a deeper understanding of the science.

First, after reading the paper thoroughly, I do not understand the use of SSRIs. The paper deals with peripheral serotonin and not brain serotonin. I can't find anything to suggest SSRIs increase peripheral serotonin. Others have even suggested there is evidence that SSRIs decrease peripheral serotonin. So, why would the researcher believe that SSRIs might help?

Second, does the persistent Type 1 IFN finding support the Itaconate Shunt Hypothesis? More specifically, am I correct in understanding that the persistent Type 1 IFN production found in this study includes IFN-a, which is what initiates the itaconate shunt pathway according to Robert Phair?

Many Thanks!

 

If I'm following, it may be the complexity of CNS vs peripheral sites of action, ie which side of the blood-brain barrier the effects are occurring, and the cross-over effects between the two.

As I understand, SSRIs have their (at least initial) effect on the serotonin made centrally in the brain, at serotonergic synapses, by reducing re-uptake at the presynaptic terminal and thereby increasing the amount of serotonin remaining and effective in the synaptic gap.

However, SSRIs also block uptake of serotonin by platelets. See Fluvoxamine: A Review of Its Mechanism of Action and Its Role in COVID-19 (2021, Frontiers in Pharmacology). Does that increase or decrease peripheral serotonin levels? ("rapidly degraded by MAOs" suggests it decreases).

So perhaps SSRIs might counter-intuitively worsen the neurological effects that might result from this paper's putative peripheral low serotonin -> reduced vagal function pathway? (If true that might shed some light on the problems of trials of SSRIs in major depression.)

---
The Expanded Biology of Serotonin (2009, Annual Reviews)

 

If I remember correctly this was to prevent LC and the time of taking the drugs was 10 days or max 2 weeks? That's not enough for an SSRI to show any real effect.

 

That's what the most of us are wondering about as well, I believe. It seems they are trying to somehow address the reduced vagal function, which they of course can't measure in a trial, and memory impairment, but of course we are more complex than a simple mice model of an acute infection. Possibly they are hoping to go after a different mechanism. These researchers are smart and quite dedicated to LC so I think the only thing we can say is that we don't know all their motivations to trial SSRIs.

This is where it gets messy. That is because they actually didn't detect any persistent type 1 IFN (they cite this study which found evidence of it, but other studies haven't and by now you'll always find a study to cite anything you want in LC). They found low serotonin (and some viral RNA in the fecal matter) of LC patients and then the rest of the study is a mouse model of the acute infection. You can be sure that they would have looked for type 1 IFN (respectively the upregulation of genes), which they found in the acute infection in mice, to see if it's happening in LC as well. Them not supplying this data means that it's very likely that there is no evidence of it happening (which is what we've seen in other studies) or that they will be publishing it in a different study (which I doubt since we've already seen newer work than this study by Peluso et al).

The study is less of a LC study then one would think. It's essentially a study of one of the pathways that explains how serotonin is lower during an acute infection with any virus. The only thing that then ties this pathway together with LC is essentially their stool RNA findings, but there is no detailed analysis of these and stool RNA samples have been negative in almost all LC studies.

One can't draw any conclusions to any other hypotheses from that (re: itaconate shunt hypothesis, the nature study they cite finds IFN-b and some other IFNs, but not IFN-a, most LC studies find IFN-y), and if one could one would also expect serotonin levels to be lower in pwME, which they aren't.