The SARS-CoV-2 spike glycoprotein interacts with MAO-B and impairs mitochondrial energetics, 2023, Pileggi et al.

Have there been any updates on this? Any opinions?

 

This paper seems related?

Background
After acute COVID-19, 5% of people experience persistent depressive symptoms and reduced cognitive function (COVID-DC). Theoretical models propose that astrogliosis is important in long COVID, but measures primarily indicative of astrogliosis have not been studied in the brain of long COVID or COVID-DC. The objective of the current study was to measure [11C]SL25.1188 total distribution volume ([11C]SL25.1188 VT), an index of monoamine oxidase B density and a marker of astrogliosis, with positron emission tomography in participants with COVID-DC and compare with healthy control participants.

Methods
In 21 COVID-DC cases and 21 healthy control participants, [11C]SL25.1188 VT was measured in the prefrontal cortex, anterior cingulate cortex, hippocampus, dorsal putamen, and ventral striatum. Depressive symptoms were measured with the Beck Depression Inventory-II, and cognitive symptoms were measured with neuropsychological tests.

Results
[11C]SL25.1188 VT was higher in participants with COVID-DC in the prefrontal cortex, anterior cingulate cortex, hippocampus, dorsal putamen, and ventral striatum than in healthy control participants. Depressive symptom severity negatively correlated with [11C]SL25.1188 VT across prioritized brain regions. More recent acute COVID-19 positively correlated with [11C]SL25.1188 VT, reflecting higher values since predominance of the Omicron variant. Exploratory analyses found greater [11C]SL25.1188 VT in the hippocampus, dorsal putamen, and ventral striatum of COVID-DC participants than control participants with a major depressive episode with no history of COVID-19, and there was no relationship to cognitive testing in prioritized regions.

Conclusions
Results strongly support the presence of monoamine oxidase B–labeled astrogliosis in COVID-DC throughout the regions assessed, although the association of greater astrogliosis with fewer symptoms raises the possibility of a protective role. The magnitude of astrogliosis in COVID-DC is greater since the emergence of the Omicron variant.

Does this potentially shed and light on disease etiology or at least potential downstream treatment?
Is anybody aware of trials or clinical experience with people with Parkinson's (or others taking MAO-B inhibitors) and Long COVID (or others with cognitive dysfunction)? Too much brain fog myself to dig deep.

What else could these results point towards if not astrogliosis/neuroinflammation?

@Jonathan Edwards do you have any insights?

 

Is there nobody here that can discuss this a bit?

 

This seems to be pretty niche and technical, so most people are probably not able to discuss it, including myself.

The authors of the paper you posted have substantial financial interests in MAO-B, so they will probably push for studies and trials if it’s worthwhile.

 

I figured, but thank you for your assessment in any case
I am wondering - as I lack knowledge/education in this field - why this line of thought linking MAO-B and LC (and mitochondrial energetics) is not looked into more.

Well spotted!

 

Most research is quite disjointed and some just work on their pet theories regardless of the merit of the theories. And most of academia is completely broken, it’s all about citations, publications, grants, power and status. I have no way of knowing if this is a part of that or not.

 

There seems to only be a handful of papers thus far examining SARS-CoV-2 interactions with mitochondria at all, so brain-specific mitochondrial interactions probably just weren't on the radar until now. Measuring that in the context of LC would be extremely challenging without good mouse models.

 

I'll preface this by saying that I'm by no means an expert on the topic here--my experience comes from spending time discussing similar cell bio papers in grad school journals clubs. So I can probably do a decent job of "translating" and can generally follow the logic, but I cannot comment on the finer details of the methodology.

Previous computational modeling has shown that the active domain of the MAO-B protein is very structurally similar to the active domain on ACE2 where SARS-CoV-2 is known to interact and use it to gain entry into cells. This made the authors think that there was a possibility of SARS-CoV-2 directly interacting with MAO-B.



Fig1 shows in-vitro (i.e. in a cell line) evidence of direct interactions between SARS-CoV-2 spike proteins and MAO-B.


For Fig2, they took a neuronal cell line and caused it to internally generate the SARS-CoV-2 spike protein. They used a particular labeling technique to measure the specific products of MAO-B enzymatic activity (14C labeling of phenylethylamine that would end up in the aldehyde product of MAO-B reactions) and found that in cells expressing the spike protein, MAO-B activity was much higher.





Fig3: Neuronal cells expressing the SARS-CoV-2 spike protein show other abnormalities in various measurements of mitochondrial function. From the text:

That last part is particularly interesting to me, since I've been investigating impaired OxPhos and compensatory glycolytic increase in ME/CFS. So this provides more evidence to existing literature potentially linking SARS-CoV-2 infection to the eventual development of bioenergetic differences associated with ME/CFS.

My main note here is that these differences in mitochondrial function cannot be solely attributed to the interaction of the spike protein with MAO-B. There could be other interactions that drive the changes beyond what was observed in Fig2. Figures 4 and 5 are a similar story--there are differences in other measures of mitochondrial function when you cause neuronal cells to express spike protein. Those changes may mediated by SARS-CoV-2 interaction with MAO-B, but they haven't proved that definitively at this step.


Fig6: Neuronal cells expressing SARS-CoV-2 spike proteins are differentially affected by MPTP in terms of cell death pathways:

The cells expressing spike protein seemed to be skewed towards necrotic cell death rather than apoptopic cell death, the latter of which tends to be mediated by mitochondria as a purposeful response (i.e. cell recognizes infection, ends up using mitochondria to kill itself instead of allowing the virus to use its machinery to replicate). This experiment is the closest that the paper comes towards showing a direct MAO-B mediated phenotypic effect since MPTP is known to be converted into neurotoxic form by MAO-B. But again, it's somewhat indirect and might be circumstantial.

My overall impression of this paper is that it provides strong evidence (with appropriate controls and other checks) to prove that:
1) the spike protein interacts with MAO-B
and
2) presence of the spike protein results in bioenergetic changes

but does not provide strong evidence that these bioenergetic changes are mediated by spike protein interaction with MAO-B. Considering that similar findings of mitochondrial dysfunction were found in other cell lines with low or non-existent MAO-B activity (which they cite), I think MAO-B likely only plays a small part of the puzzle.

Furthermore, they're using only the spike protein, not the whole virus. There's a reason for doing this (if they used the whole virus, they'd probably just see a lot of cell death). This study, along with others, provides evidence that just the spike protein is enough to cause (at least short term) changes in mitochondrial function, but there might be more to the story mediated by the rest of the virus (e.g. other viruses are known to produce additional proteins that directly interact with host proteins, so you'd only see that interaction if you were infecting cells with more than the spike protein).

 

So in blunt terms, the cells are murdered rather than committing suicide?

 

Kind of--it's less murder and more death by neglect, since it wouldn't be good for the virus if the cell dies prematurely by suicide or murder. Necrotic death usually comes from an accumulation of toxins when the cell can't function properly. It's more like the virus prevents the cell from using its own kill switch, keeps it alive long enough to co-opt all the machinery for its own purposes, and then leaves it for dead (assuming that the cell wouldn't just die anyways when the virus replicates and blows a hole in the cell membrane to spread further).

Something like the Alien movies I suppose. Which I guess could be viewed as murder, but that's more of a semantic distinction haha

 

Thank you for explaining, that makes sense to me!

 

One way this paper could have been improved is by knocking down MAO-B in the cells and seeing if the observed bioenergetic changes still occur when spike protein is expressed. If you only see the mitochondrial changes in the presence of MAO-B + spike protein, this would provide evidence that all the changes are directly mediated by MAO-B and not by some other interaction between the spike protein and mitochondria. From looking at the literature, it seems like MAO-B is not vital to neuronal cell survival, so knocking it down would be viable.

 

The cynic in me wonders if this was not done on purpose. You wouldn’t run the risk of disproving your pet theory, especially when you have a financial interest.

 

Sorry I missed this.
I cannot get excited about any of this being of any relevance to ME/CFS or Long Covid much. We see similar illness following lots of infections so individual antigens and their similarity to human proteins are likely to be irrelevant, in my view.

This looks to me like the typical tunnel vision of much biomedical science these days. Measure anything you can measure and forget the fact that there are a lot of context factors that make most theories implausible before you even start.

Neurons may well die in people on intensive care during acute Covid but I doubt neuronal death has anything to do with LC - or astrogliosis.

 

Did they show that it's something specific to spike? I mean maybe if they made the cell produce any other random foreign protein, like a piece of influenza, the same changes might happen, right?

 

They only tested SARS-CoV-2 spike so it absolutely could be something that other viruses can also do.

 

Interesting. It does seem though, that MAO-B is more active in both papers, one being a human trial? I'm not very familiar with this though, so I'm wondering what else (besides their ideas about astrogliosis) could be reason for getting such a result (methodologically or pathophysiologically)?
In this case, what context factors make this theory implausible?