SARS-CoV-2 spike protein induces TLR4-mediated long-term cognitive dysfunction recapitulating post-COVID syndrome in mice, 2023

[Mij]

Authors:
Fabricia L. Fontes-Dantas et al

Highlights

• Spike protein infusion into mouse brain induces late cognitive dysfunction
• Spike protein induces late hippocampal microgliosis and synapse loss
• Blockage of TLR4 renders mice resistant to Spike-induced cognitive dysfunction
• TLR4-2604G>A GG genotype was related to poor cognitive outcomes in COVID-19 patients

Summary
Cognitive dysfunction is often reported in patients with post-COVID, but its underlying mechanisms are not completely understood. Evidence suggests that SARS-CoV-2 Spike protein or its fragments are released from the cells during infection, reaching different tissues, including the CNS, irrespective of the presence of the viral RNA.

Here, we demonstrate that brain infusion of Spike protein in mice has a late impact on cognitive function, recapitulating post-COVID syndrome. We also show that neuroinflammation and hippocampal microgliosis mediates Spike-induced memory dysfunction via complement-dependent engulfment of synapses. Genetic or pharmacological blockage of TLR4 signaling protects animals against synapse elimination and memory dysfunction induced by Spike brain infusion.

Accordingly, in a cohort of 86 patients recovered from mild COVID-19, the genotype GG TLR4 -2604G>A (rs10759931) is associated with poor cognitive outcome.

These results identify TLR4 as a key target to investigate the long-term cognitive dysfunction after COVID infection both in humans and rodents.

https://www.cell.com/cell-reports/fulltext/S2211-1247(23)00200-0

 

[Hutan]

Brazilian team

Introduction

There are evidence suggesting that during SARS-CoV-2 infection, Spike protein or its S1 fragment are released from the cells, reaching different tissues, including the CNS, irrespective of the presence of the viral RNA 15,16. Additionally, it has been demonstrated that cells expressing the Spike protein release extracellular vesicles containing the full‐length protein 17, which would be another means of its circulation in the body. Free S1 was shown to cross the blood-brain-barrier (BBB), reaching different memory-related regions of the brain, suggesting that the protein itself, independently of the viral particles, would affect brain functions 18. Notably, Swank and colleagues (2022) detected high levels of circulating Spike protein several months after SARS-CoV-2 infection in patients diagnosed with post-COVID, but not in the individuals that did not present long term sequelae.

There does seem to be evidence that virus particles do get into the brain, and for some time after the acute infection.

TLRs are activated by different pathogen-associated molecular patterns (PAMPs) and are crucial for evoking the innate immune responses to infection, stress or injury 20. Studies have predicted that SARS-CoV-2 Spike protein binds to TLR4 with higher affinity than it binds to ACE2 21,22, and its aberrant signaling is involved in the hyperinflammatory response of patients with COVID-19 23. In vitro studies also demonstrated that SARS-CoV-2 Spike protein activates TLR4 in cultured phagocytic cells, stimulating production of proinflammatory mediators.

The paper did not say in the abstract or Introduction, but I'm assuming TLRs is Toll-like receptors. So, they are suggesting there is evidence for the CoV-2 spike protein binding to TLR4, which then produces an inflammatory response.


Results - injection of spike protein into mice
Temporary reduction in cognitive performance

While the vehicle-infused mice (Veh) were able to perform the NOR task, as demonstrated by a longer exploration of the novel object over the familiar one (Figs. 1B-E, white bars), mice infused with Spike failed to recognize the novel object when evaluated between 30 and 45 days after injection (Fig. 1C, D, black bar). Remarkably, memory dysfunction is a late outcome of brain exposure to Spike protein as at the early time point (7 days after infusion), the 7 animals were still able to perform the NOR task (Fig. 1B, gray bar). Of note, performance of i.c.v. Spike protein-infused mice in NOR test returned to normal at 60 days after infusion (Fig. 1E), showing that memory impairment is reversible.

Mice infused with the CoV-2 protein performed ok on what is described as a cognitive task 6 or 7 days after infusion (See Figure 1B), but didn't later (Figures 1C, D), whereas the control mice performed ok throughout. They say that by 60 days (1E), the spike protein treated mice had regained their cognitive ability.



The authors suggest that this is a pure cognition test, rather than related to a general sickness behaviour, as measures such as time spent exploring the objects weren't different. The results do seem to reflect what they report; the asterisks indicate significant differences between the time spent exploring novel objects versus familiar objects ( in charts B and E for the spike protein treatment, in all the charts for the control treatment).

They also administered the spike protein subcutaneously (rather than direct into the brain) and report that the results of the memory test were similar. They also tested the mice with a water maze; they report similar results.


Reduction in synaptic density
They looked to see if there was a loss of synaptic density in the hippocampus. They didn't find any differences compared to the controls in the early stages (Fig 1J-N), but did later (Fig 1O-S). They found that synaptic loss was not caused by neuron cell death. I don't know what co-localised puncta is, but there does seem to be a significant difference between the result for the treated group at the early time (N) and the later time (S).



I'm going to take a break. There is a lot in this paper. Of course a big question is 'why do the effects we see in Long Covid often persist beyond the 60 days when the mice in these experiments seem to come right during that time?'.

I should really read the whole paper before I start to post about it, to produce a coherent summary but I suspect that if I did that, I wouldn't end up posting much. So, you get a sort of stream of consciousness, a record of what I find interesting as I read through.

 

[SNT Gatchaman]

The paper did not say in the abstract or Introduction, but I'm assuming TLRs is Toll-like receptors.

Review article: Toll-like Receptors and the Control of Immunity (2020, Cell)

 

[Sisyphus]

One has to wonder if the spike protein produced after mRNA covid vaccine has the same effect.

 

[Hutan]

One has to wonder if the spike protein produced after mRNA covid vaccine has the same effect.

I suppose potentially it might. The researchers tried a lower dose of the protein (10-fold smaller), and it didn't affect memory at all, so dose seems important. I haven't looked at and thought about what dose they used, but it's possible that the dosage used in this study in mice is not a reasonable model for what happens in humans during a Covid-19 infection except in the most extreme of circumstances.

If a noticeable impact on cognition was a major risk from vaccination, I think we would have heard, even if, as here in mice, it was a fairly short-lived effect.

 

[Hutan]

The spike protein triggers neuroinflammation
The researchers tried applying the protein directly to the neutrons - that didn't seem to injure the neurons or affect synaptic density.

In the early stage after injection of the spike protein into the brain didn't change the number or shape of microglia, or increased expression of a range of inflammatory genes (TNF, IL-1b, IL-6, INF-b, IFNAR1) but levels of IFNAR2 mRNA did decrease. At the late stage, mRNA levels of TNF, IL-1b, IFNa and IFNb and IFNAR2 increased in the hippocampus. Protein levels of TNF and IL-1b were higher in the hippocampus too. Hippocampal expression of IL-6 and IFN-y cytokines and the receptor IFNAR1 were not affected. They also found increased serum levels of TNF at the late stage which returned to normal by day 60.

They looked at sections of the hippocampus to see if there was evidence of gliosis (glial cell types include microglia and astrocytes; gliosis is when glial cells proliferate in injured areas of the brain). There was no evidence of microgliosis at the early stage, but, at the late stage, they found an increased number of Iba-1-positive cells (Fig. 2K-M) and a predominance of cells with ameboid morphology in the hippocampus (Fig. 2K,L, N), as well as significantly higher TMEM119 immunoreactivity in the DG hippocampal subregion of Spike-infused mice (Supplementary Fig. 5N-P).

They concluded that the cognitive impairment that they believe happens after infusion of the spike protein is accompanied by microglial activation and neuroinflammation.

Synaptic phagocytosis by microglia
They found evidence that activated microglia are pruning off and internalising synapse, as is known to occur in dementia and viral encephalitis.

This is Figure 3A, 3B. The images are from the late stage. In 3A, from mice hippocampus tissue treated with the control infusion, the synapse is intact, and the microglia is branched and looking normal. In 3b, which had the spike infusion, it is said that synapse bits are inside a microglia. Yeah, if you squint it's plausible.


Here are measures of co-location of the synapse and microglia markers in two different parts of the hippocampus (C and D). They also looked at concentrations of complement 1q (C1q), which they say is involved in the tagging of synapses for engulfment by the microglia. They found no increase of C1q at the early stage, but a big increase at the late stage.



So, so far, so good. But it's not terribly surprising that brains react when you inject a pathogen protein into them, or even that cognitive function is associated with that. The big news so far, I think, is that sub-dermal infusion can possibly cause the same issues, that the pathogen protein crosses the BBB. And still we don't know if the problem could possibly become chronic (although this paper might not address that).

It might worth looking to see if there are any ME/CFS studies relevant to these findings - I'm not sure how much microglial activation has been found.

 

[Hutan]

Just wanted to park this diagram here, as it shows how a pathogen or pathogenic material (the LPS at the top left) can impact TLR4 (mentioned in this paper), and how that could then go on to activate the itaconate shunt, down regulating glycolysis.

Source: The role of itaconate in host defense and inflammation

 

[rvallee]

No reason to think this is unique to this particular protein? Should be true of other proteins and molecules recognized as foe by the immune system. If you inject directly into the brain anyway.

So the issue would be more about the mechanism by which viral proteins can reach that far, particular or even unique to the pathogen, but activating the immune system with molecules that trigger TLRs probably have similar effects.