The spike protein from the COVID vaccines is found in humans for at least four months after injection. It is a direct toxin that pokes holes in cell membranes. It causes mitochondrial damage in cells and COVID-like illness in mice.
But is it found in humans at concentrations that will create these problems? And are the modifications made to the vaccine spike proteins likely to reduce this toxicity?
This is the first in a series of posts in which I will outline the potential mechanisms of spike protein toxicity and vaccine side effects and brainstorm actionable steps vaccinees can take to protect against these effects or heal from them.
In this first post in the series, we look at the spike protein as a pore-forming toxin. First, some essential background.
Background: ACE2 in Normal Physiology and in Viral Infection
The spike protein of SARS-CoV-2, the virus that causes COVID-19, is the spike-shaped protein that sticks out from the viral membrane and is responsible for its ability to bind to, fuse with, and enter human cells.
It does this by binding to ACE2. From the "perspective" of the virus, ACE2 is a "receptor" for cell entry. Its best-understood role in normal human physiology, however, is as a zinc-dependent enzyme present on the cell surface that breaks down or prevents the formation of angiotensin II in the blood. In doing so, it produces angiotensin(1-7), which balances the actions of angiotensin II. Angiontensin II constricts blood vessels and raises blood pressure. In excess, it creates scar tissue, also known as fibrosis, especially in the lungs, blood vessels, and heart. By decreasing angiotensin II and increasing its counterbalance, angiotensin(1-7), ACE2 promotes healthy blood pressure and keeps the tissues of the lungs, blood vessels, and heart in healthy shape.
While human health usually suffers from too little ACE2 rather than too much, animal experiments suggest that ACE2 has its own dark side: too much disturbs the rhythm of the heart and causes sudden death.
While most ACE2 is embedded in the cellular membrane and bound to the cell surface, some is secreted from the cell into the blood. This freely circulating ACE2 is thought to act as a decoy for certain viruses by preventing them from binding to the surface of cells they could otherwise infect.
Background: Spike Protein Subunits and "S1 Shedding"
Like the spike proteins of other coronaviruses before it, the COVID spike is composed of several subunits.
S1 contains the receptor-binding domain (RBD), the part that binds to its "receptor," ACE2, on the cell surface. This causes the virus to stick to the cell it is about to invade, but isn't enough to cause the membranes of the cell and virus to fuse with one another or to allow the virus to enter the cell.
S2 is composed of two parts, S2' and the fusion peptide. The fusion peptide is what causes the viral membrane to fuse with the human cell membrane, allowing other mechanisms of the cell to take over and enable entry of the virus into the cell.
The infamous "furin cleavage site" is a part of the spike protein targeted by the human enzyme furin, which cleaves S1 and S2. This in and of itself does not cause S1 to break away. It also is not necessary for the binding of the S1 unit to ACE2. However, it does increase ACE2 binding. This is because each spike protein complex has three S1 units with the ability to bind to ACE2. Before cleavage, usually only one of them is open and available for ACE2 binding. After cleavage, the spike is more likely to open up and bind one or two more ACE2 molecules.
However, cleavage does not always increase ACE2 binding. As discussed toward the end, the J&J vaccine has the cleavage site blocked, yet displays increased ACE2 binding.
Cleavage of S1 from S2 is also thought to be necessary to expose the S2 subunit to a second cleavage site that breaks it into S2' and the fusion peptide. By contributing to the release of the fusion peptide, therefore, both cleavages appear to contribute to cell fusion.
These cleavages increase the likelihood the spike protein will change shape, from a "prefusion" to a "postfusion" conformation. This involves a "jackknife" maneuver where the fusion peptide does a backflip and sticks itself into the cell membrane. This change of shape appears to be the most necessary part of the whole process for cell fusion. The cleavages appear to drive it, although it can also happen prematurely while the spike proteins are first being produced within cells.
In a poorly characterized process of "S1 shedding," at least some S1 is released, both bound to soluble ACE2, and as free, unbound S1.1
The Spike Protein is a "Pore-Forming Toxin" that Pokes Holes in Cell Membranes
The free S1 subunit of the spike protein is a toxin that destabilizes and disrupts lipid membranes without the need to interact with any receptors or other proteins. This is shown by making synthetic lipid bilayers permeable to charged ions. This suggests that it pokes holes in lipid membranes, making it a "pore-forming toxin."
The membrane is the principal form of cellular organization and its integrity is central to the cell's ability to produce energy, survive, and thrive. As such, it is difficult to understate how all-encompassing this type of toxicity is. For more background on the importance of membrane integrity, see the footnote.2
With well characterized pore-forming toxins we know details of exactly how the pores form, how big they are, and what size substances they do and do not let through. We do not have these details for the spike protein. We only know that its subunits make membranes permeable to charged ions they would otherwise not let through.
This effect occurs in synthetic lipid membranes at concentrations at least as low as 40 nanomolar (nM, a measure of the concentration of molecules).
Within human lung cancer cells, the concentration required to kill cells starts somewhere between 1 and 2 micromolar (μM), which is 1000-2000 nM, and 25-50 times greater than 40 nM. 24-hour exposure to 2 μM kills off 30% of the cells.
In in vitro models of the blood-brain barrier (BBB) using human cells, both S1 and S2 disrupt the integrity of the barrier in concentrations as low as 0.1 nM.
This BBB paper also showed several other things:
- 50 nM S1 causes a loss of proteins that form the junctions between the cells.
- 10 nM S1, S2, or RBD cause the cells to make ICAM-1 and VCAM-1, which are two "adhesion molecules." The production of adhesion molecules can be seen as the cells crying out to the immune system that they need help. That is, these are molecules that will cause immune cells to migrate to the BBB.
- Consistent with these cries for help, the same treatments cause increases in inflammatory cytokines and inflammatory zinc-dependent enzymes known as MMPs that break down tissues.
However, they showed this breach of integrity starting at 100 times lower concentration (0.1 nM) than they used to show a pro-inflammatory response (10 nM), and at 500 times lower concentration than they used to cause loss of proteins at cell junctions (50 nM).
This paper showed a "breach" in the barrier function of the BBB by showing that it became permeable to small, charged molecules. This is very similar to what the first paper had shown in synthetic lipid membranes. Yet those synthetic membranes did not have any ability to initiate an inflammatory response and did not have any proteins forming junctions. There were no immune cells involved. Thus, the findings of the BBB paper are at least as attributable to the spike protein acting as a direct, pore-forming toxin, poking holes in the cell membrane, as they are to the inflammatory response destroying the junctions between the cells.
The findings of these papers are consistent with a minor finding in a subsection of a mouse paper discussed below, where S1 at concentrations at least as low as 10 nM disrupted the barrier formed between primary human lung microvascular endothelial cells. These are the cells that make up the lining of the small blood vessels of the lungs, which are part of the gas exchange machinery.
Taken together, these papers suggest the following:
- S1 and S2 probably both act as direct, pore-forming toxins at concentrations as low as 0.1 nM.
- Concentrations at least as low as 10 nM elicit an inflammatory response.
- The inflammatory response creates a second layer of destruction that compromises the junctions between cells.
The pore-forming toxin effect would likely create all sorts of cellular destruction everywhere, since no cell can function at all without healthy membranes.
Cellular junctions are important everywhere, but their disruption would be particularly damaging to tissues that form important barriers. These would especially include the skin, blood vessels, blood-brain barrier, and the mucous membranes of the eyes, nose, mouth, throat, respiratory tract, and gut. It would also disrupt the function of fibers that rely on such connections to act in unison, such as the contractile units of the heart responsible for its rhythm.
The Spike Protein Causes COVID-Like Illness and Lung Damage in Animals and Fragments Their Mitochondria
In mice, direct injection of 10 micrograms3 (μg) of S1 into the trachea, blown with air into the lungs, causes COVID-like illness, COVID-like lung damage, and a COVID-like cytokine storm across the span of three days.
A similar but lower-quality study in hamsters came to a similar conclusion.4
The hamster paper also showed that in isolated cells roughly 50 μM of S1 will fragment mitochondria, hurt their ATP production, and make them generate more acidity, apparently by shifting them toward anaerobic glycolysis so that they generate more lactic acid.
The mouse study is particularly remarkable because it raises the possibility that a portion of the PCR-negative COVID-like illness could be spike protein toxicity caused by the vaccines. This could explain much of the Hospitalization Paradox, since it provides a mechanism whereby the vaccines can replicate the disease without any chance of generating a positive PCR test.
However, to really make the case that the vaccines could cause spike protein toxicity, we need to take a closer look at the concentrations of spike protein the vaccines produce, and modifications the vaccines make to the spike protein.
First, let's look at another known pore-forming toxin to see if pore-forming toxins can, in general, cause the effects described above in cells and animals.
Other Pore-Forming Toxins Also Cause Pneumonia, Lung Damage, Myocarditis, and Mitochondrial Fragmentation
There are some remarkable similarities between the spike protein and another pore-forming toxin, pneumolysin, produced by Streptococcus pneumoniae, the major bacterium responsible for community-acquired pneumonia.
As with the spike protein, direct application of pneumolysin to the lungs of mice causes the major features of lung injury that occur in bacterial pneumonia, including edema (swelling) in the lungs, leakage of the blood vessels, and pulmonary hypertension.
The quickest thing to occur is a disruption of barrier function in the lungs and their blood vessels, beginning within 30 minutes. Hypertension appears to result from the formation of protein fibers that make the blood vessels stiffer as compensation for their disrupted membranes.
Infiltration of the lungs with immune cells known as neutrophils does not happen appreciably until the 24-hour mark. This suggests that the primary damage occurs as a result of the toxin, rather than the immune response. No bacteria need be present.
Remarkably, these findings suggest that pneumonia, whether bacterial or viral, may be better described as a "poisoning" than an "infection," with the infection simply serving as the usual delivery vehicle for the poison. (Until, perhaps, you develop another vehicle, such as an injection of the DNA or mRNA required to make the toxin.)
Like the spike protein, pneumolysin impairs ATP production, fragments mitochondria, and kills cells.
Holes in the mitochondrial membranes allow hydrogen ions to flow through them indiscriminately so that the mitochondria cannot effectively harness their flow to make ATP. (see footnote 2 for background). These holes also cause calcium to leak into the mitochondrion, which serves as an emergency signal that elicits a cell suicide response known as apoptosis.
As with COVID and the COVID vaccines, bacterial pneumonia can cause myocarditis. While I have not found any studies showing that pneumolysin can cause myocarditis on its own, in at least one strain of the bacterium a mutation causing pneumolysin to be deficient abolishes its ability to damage the heart. This shows that pneumolysin can be a contributor to myocarditis.
Altogether, these studies suggest that impaired ATP production, mitochondrial fragmentation, barrier dysfunction, myocarditis, pneumonia, and acute lung injury may be general features of pore-forming toxins, providing they reach the right tissues in sufficient concentrations. Thus, the apparent pore-forming nature of the spike protein at low concentrations should be a leading candidate for explaining the harm of both COVID and COVID vaccines.
Does Vaccine Spike Protein Reach High Enough Concentrations to Be Toxic?
First, we turn our attention to the concentration of spike protein we should expect the vaccines to produce.
We have five metrics:
- The pore-forming toxin role appears to start at least as low as 0.1 nM.
- Immune responses are elicited at least as low as 10 nM.
- Rampant cell death occurs within 24 hours when concentrations are between 1 and 2 μM (between 1000 and 2000 nM).
- Mitochondrial impairment and fragmentation occur at least as low as 50 μM.
- COVID-like illness, lung damage, and cytokine storm occur when 10 μg of S1 reach the lungs of mice.
These concentrations are well above the minimal concentrations shown to disrupt barrier function, presumably by acting as a pore-forming toxin.
They fall short of the other targets.
However, the vaccines obviously elicit an immune response, making the 10 nM target questionable.
The rampant dieoff of cells in 24 hours of exposure is clearly very far from being met, and that is also obviously not happening since most people who get the vaccine would be dying on day one if it were.
They fall even wildly shorter of the 50 μM used to fragment mitochondria and shift them toward anaerobic glycolysis. However, this almost certainly occurs at much lower concentrations if the cells used in this experiment had similar sensitivity as those used in the one mentioned above. Mitochondrial fragmentation would be expected to cause cell death and be an indicator of cell death, as occurs with pneumolysin. Thus, this almost certainly can happen at 2 μM.
If, moreover, the impairment in mitochondrial function is simply a result of the spike protein acting as a pore-forming toxin, then we can expect the concentrations reached to at least mildly impair mitochondrial function even if far less dramatically than observed in that study. This, perhaps, could contribute to fatigue, electrolyte imbalances, twitching, exercise intolerance, blood sugar problems, poor healing, unhealthy aging, or other such dysfunctions rather than death.
We do not know the concentrations reached within cells, but they are presumably higher than those in the blood, given that spike protein is first produced inside cells and only some portion of it should be released into the blood.
Now let's compare the 10 μg of S1 that causes COVID-like illness in mice when injected into their trachea and blown into their lungs.
We don't know how much spike protein is produced from the vaccine or where it winds up. However, my extremely rough back-of-the-envelope math in the footnote 5 suggests that the amount of spike protein we would expect to reach human lungs after a single dose of an mRNA vaccine falls short of the human-adjusted equivalent of what was used in the mouse study by anywhere from 1000 to 12,000-fold.
However, these calculations should be taken with a grain of salt for a few reasons:
- They are based on extremely poor substitutes for data that simply do not exist. (That's right, we still have no biodistribution data for actual spike mRNA or protein.)
- We don't know the comparative sensitivity of animal and human lungs to spike toxicity.
- There is likely variation from batch to batch and syringe to syringe of vaccine.
- There could be lots of variation from person to person in where the mRNA goes and how much protein it makes.
- If some injections accidentally hit a vein, far more mRNA goes everywhere it's not supposed to.
- Finally, the huge, giant wildcard standing in the room like an elephant: the recent finding that the mRNA can be reverse transcribed into DNA may mean far more spike protein gets produced than we otherwise would have thought. That is, if each mRNA molecule makes 1,000 copies of itself through conversion to DNA and back, suddenly the amount of spike protein reaching human lung enters the range where extrapolation from the mouse study becomes plausible.
Therefore, we can say the following about the concentrations of spike protein reached:
- The spike protein definitely reaches high enough concentrations after vaccination to disrupt lipid membranes, most likely by acting as a pore-forming toxin.
- This is probably high enough to contribute to some impairment of mitochondrial function.
- It is plausible but very uncertain that the spike protein may sometimes reach high enough concentrations in human lungs to cause illness, lung damage, and a cytokine storm that all resemble what occurs in a severe COVID case.
The Pfizer, Moderna, and AstraZeneca vaccines all code for spike proteins that retain their native furin cleavage site, allowing the splitting of the protein into the S1 and S2 subunits. J&J is the one vaccine that knocks out the furin cleavage site.
The S1 subunit is the part that binds to ACE2, but the furin cleavage site is not needed to bind to ACE2. This is quite clear from the fact that the block of the furin cleavage site in the J&J vaccine actually increases its ability to bind to ACE2 compared to the natural spike, yet stops the S1 subunit from "shedding" off the surface of the cells.6
All but AstraZeneca have been "prefusion stabilized." This keeps the protein from jackknifing and sticking its fusion peptide into the cell. However, the vaccine spike protein is not part of a viral membrane and there is no viral membrane that is trying to fuse with the cell membrane, so any restriction this imposes on the fusion peptide is irrelevant.
The main reason this is done is to improve the immune response. In the transition to the postfusion form, sugars move from the inside to the outside of the spike protein, thereby creating a sugar coating that helps it evade the immune system. Moreover, an immune response to the postfusion form is thought to be directed at the spike protein too late in the process to intervene with cell entry. Therefore, the prefusion stabilization increases the immune response and is thought to make the immune response more productive.
The two mutations used for prefusion stabilization had originally been developed during efforts to develop vaccines for earlier coronaviruses, HKU1, MERS, and SARS. They don't stop binding to ACE2, they just stop the change of shape that happens afterwards. In fact, in the context of the SARS virus, the prefusion stabilized spike bound to ACE2 slightly more strongly than the natural spike. The binding strength reported for the Pfizer vaccine is stronger than that reported for the natural SARS-CoV-2 virus.7 Experiments in the development of the J&J vaccine showed that prefusion stabilization increased ACE2 binding.8
S1 shedding increases toxicity, and may be necessary for it.
The role of ACE2 binding is unclear. The cell experiments discussed above used isolated S1 or S2 subunits rather than whole spike proteins.
The mouse paper showed that whole spike protein induced much more mild membrane disruption within human cells and much more mild lung injury in live mice than the S1 subunit. At a mimimum, this suggests that S1 shedding worsens toxicity. Since nothing was done to prevent the mice or cells from cleaving the whole spike protein, moreover, it may be that S1 shedding is an absolute requirement for toxicity.
If that is the case, it may be that the J&J vaccine is less culpable for pore-forming toxicity than the others, or even immune to it entirely. We should be careful, however, because the necessity of S1 shedding for toxicity is not yet certain, and because it is possible that certain tissues possess enzymes other than furin that could cleave even the J&J spike protein.
The mouse paper showed that genetically engineering the mice to make the human form of ACE2 caused the lung injury to be much worse. Mouse ACE2 does not bind to the spike protein of SARS-CoV-2 effectively, but human ACE2 does. Mice that are engineered to make human ACE2 make just as much of the mouse version of ACE2, but also make ten times more human ACE2 over and above that. That gives them no deficit at all of the normal mouse ACE2 but roughly eleven times as much total ACE2.
These mice are usually used for SARS or COVID experiments because it allows the virus to enter the cells, which mice are otherwise resistant to.
However, free S1 subunits have no S2, no fusion peptide, and no viral membrane to fuse. What is ACE2 binding supposed to enable?
Overexpression of human ACE2 in mouse heart causes arrhythmia and sudden death. This appears to be a result of deficient angiotensin II. While too much angiotensin II is bad, some is needed to form the junctions between the cells. Overexpression of ACE2 in mouse heart impairs the formation of those junctions. If there is a similar effect in the lungs, it may simply be that the pore-forming toxicity of the spike protein creates more lung damage when the lungs already have defective barrier function due to overexpression of ACE2.
Regardless, the vaccine spike proteins bind ACE2 at least as well as the spike of the natural virus, so if there is a role for ACE2 binding, they are not absolved. Overall, then, the only modification that might interfere with the pore-forming toxicity of the spike protein is the furin cleavage site blockage in the J&J vaccine.
Vaccine Spike vs Natural Spike
The pore-forming toxicity of the natural spike may indeed be of concern, and may even be a leading candidate to explain much of the lung damage that occurs in natural infection, just as the pore-forming toxin of Streptococcus pneumoniae, pneumolysin, explains the lung damage of bacterial pneumonia.
However, the vaccine spike guarantees the systemic distribution of this pore-forming toxin and raises the possibility that membrane disruption, mitochondrial damage, and barrier dysfunction could reach far and wide, outside the bounds of the respiratory tract. We know from the Pfizer documents that the liver, spleen, ovaries, bone marrow, adrenal glands, and intestines are particularly concerning.
The mRNA of the spike protein is highly modified to make it evade the innate immune system, last for much longer, and be translated into protein at a much higher rate.
By contrast, the natural virus for most people will be limited to the respiratory tract, and to a lesser extent the eyes or the gut. Virus is only found in the blood of 44% of those on a ventilator, 27% of those hospitalized, and 13% of those treated as outpatients. As I pointed out in Explaining the Hospitalization Paradox, "Almost certainly the incidence is even lower in those with mild cases who never seek hospital treatment, and it is almost certainly non-existent in those who were exposed without ever feeling ill." Where it is encountered, it is more easily destroyed by the immune system. Hence, the spike protein is consistently found in the lymph nodes of the arm pits at least 60 days after vaccination, but traces are rarely found at all after natural infection.
The Bottom Line
While there are other mechanisms of spike protein toxicity I will write about, the most compelling mechanism that occurs at the lowest concentration is the disruption of cellular membranes.
The spike protein apparently acts as a pore-forming toxin that pokes holes in cell membranes at concentrations well below those found circulating after vaccination.
This may explain COVID-like illness, pneumonia, lung damage, myocarditis, mitochondrial dysfunction, broad-based cellular dysfunction, and barrier disruption in the blood vessels, blood-brain barrier, skin, lungs, and gut.
This mechanism has not been definitively shown to cause vaccine side effects, but it should be considered a leading candidate to explain them.
This is a concern for both the natural spike in the virus and for the vaccines, but the vaccines pose a much greater risk of systemic distribution and persistence over time.
The J&J vaccine may be exempt from this specific mechanism of toxicity, or at least pose a lower risk of it, but this needs further study.
In future posts in this series, I will cover other mechanisms of spike protein toxicity as well as other mechanisms of vaccine side effects that do not involve direct toxicity of the spike protein, and will brainstorm actionable steps vaccinees can take to protect against these effects or heal from them.
Comment: See also: