Snake venom and Covid

Snake venom is in Remdesivir, snake venom is in our water supply, andCovid-19 is really snake venom poisoning. This social media debacle is part of a web of misunderstanding.

1. Dr Brian Ardis was presented with a pro- therapeutic antibody therapy for Covid-19 metaphor. Snake bites are treated with antibodies against venom toxins.
2. Lay people came across a “peer” reviewed publication in a lesser journal suggesting the presence of a huge list of snake venoms in blood plasma and feces of Covid-19 patients.
3. A University of Arizona professor discovered that we humans secrete our own soluble phospholipase A₂ (PLA₂) in Covid-19 infections. PLA₂ is a very common component of snake venoms. This is why Covid-19 infections have similar symptoms to snake bites.

this post will try to establish clarity to a social media $hit storm.

This post is not going to dissect nor criticize ramblings of non scientists speaking the wanderings of their minds. This Stew Peters show wanders from one conspiracy theory to another.

1. What is a metaphor?

Remember in high school English when we learned about metaphors and similes? For example:

“Delia was an overbearing cake with condescending frosting, and frankly, I was on a diet.” Maggie Stiefvater. Stiefvater could simply have said, “Maggie was like an overbearing cake..” We all know that the story is not about a cake, right? Stiefvater does not have to use the word “like” for us to know this. Perhaps the misunderstanding would not have occurred if the narrative was, “Covid-19 is like snake venom, just take the beeping monoclonal antibody, dumb ass!”

Then they start going off on monoclonal antibody therapy for snake bites.The reasoning is that if FDA, CDC, NIH tell us to do something, we should do the opposite. Of course we’d use anti venom antibodies with a snake bite.

2. The snake venom paper that started it all

Brogna C, Cristoni S, Petrillo M, Querci M, Piazza O, Van den Eede G. Toxin-like peptides in plasma, urine and faecal samples from COVID-19 patients. F1000 Res. 2021 Jul 8;10:550. doi: 10.12688/f1000research.54306.2. PMC free article

This paper used Liquid Chromatography-Surface Activated Chemical Ionization – Cloud Ion Mobility Mass Spectrometry (LC-SACI-CIMS). They are basically using a technique to impart a charge to a molecule and measuring the mass/charge ratio. When using mass spectrometry to sequence “peptides”, one usually finds a way to blast them to little pieces and then measures the mass/charge of the little pieces. Usually whole proteins are digested into discrete peptides with a protease like trypsin or chymotrypsin. These proteases cleave at discrete amino acid sequences. Then the peptides are blasted into little pieces of discrete masses. If they were doing this, it would be LC-SACI-CIMS/MS. Collision induced dissociation (CID) seems to be the second part. Then they searched a venom specific database for possible fragments. Why? Normally, when dealing with human samples, one searches a database containing all human proteins. One might add a database of Covid proteins too for good measure. Why a library of all venom proteins?

Brogna and woworkers used the following technique for removing peptides from human blood plasma: “5 µL of CH ₃CN were added to 50 µL of plasma and vortexed for one minute. The procedure was repeated 10 times. Then the sample was centrifuged at 1,500 g for 10 minutes and two 100 µL aliquots of supernatant were dried and resuspended in 70 µL of NH ₄HCO ₃ 50mmol.” One thing they did not do was add a reducing agent to break about disulfide bonds.

Figure 3, what about the disulfide bond absolute of snake venoms?

In Figure 3 try to convince us that Covid blood plasma contains conotoxin from the virgin cone snake. The top portion of this figure is from UniProt. According to UniProt, this processed protein contains four disulfide bonds. The sequence is shown. Arrows point to the cysteines that participate in the disulfide bonds. The structure of cysteine is shown in the upper right hand. The “S” is a sulfur that may form a disulfide bond with other cysteine sulfurs in the venom. The cartoon is a whimsical statement of the importance of disulfide in venom structure.

The bottom half of this figure is the Bogna figure 3.

The bottom half is their figure 3. Arrows pointing to cysteines have been added. Unless there was a typo that the reviewers missed, three is no way these masses can be correct because of the disulfide bonds. The Bogna panel 3b is not shown because it simply explains their source of CID fragment ions without even addressing the disulfide issue. This image was obtained with an Internet. When disulfide bonds are not reduced, scientists are required to show expected fragmentation patterns.

Figure 4, the whole conotoxin protein

The top half is data from the Bogna publication. The bottom half was taken from UniProt. Bogna and coworkers claim to have found fragments from the entire protein, not just the processed toxin.

1. The signal peptide is there to direct the newly synthesized protein to secretion pathways. It is usually cleaved off before secretion. One would expect this peptide to stay in venom gland cells of the snake. Even if these peptides are synthesized artificially, why bother making something that is useless in terms of poisoning?
2. The propeptide is there to keep the toxin inactive so that it does not poison the snake. Perhaps some venom protease cleaves it off and then maybe we’d expect to see it in the body fluids of a patient. If it is there, it might have come from an actual snake. As for conspiracy theories of synthetic production: Why bother?
3. The processed toxin contains many disulfide bonds. It would be difficult to produce a synthetic peptide with all of the disulfide bonds in the right place.

The authors claimed that they found these peptides in all five of the plasma samples and three of the fecal samples. There are several things about these data.

To their credit, these authors never suggested any conspiracy theories or that Coivd-19 patients were being secretly injected with snake venom in therapeutics like Remdesivir.

3. We secrete PLA2 in response to Covid

Snider, J. M., You, J. K., Wang, X., Snider, A. J., Hallmark, B., Zec, M. M., Seeds, M. C., Sergeant, S., Johnstone, L., Wang, Q., Sprissler, R., Carr, T. F., Lutrick, K., Parthasarathy, S., Bime, C., Zhang, H. H., Luberto, C., Kew, R. R., Hannun, Y. A., Guerra, S., … Chilton, F. H. (2021). Group IIA secreted phospholipase A2 is associated with the pathobiology leading to COVID-19 mortality. The Journal of clinical investigation, 131(19), e149236. Free article

The scientific study had to be tracked down on PubMed. This study is truly hard core and probably too much for the average reporter that knows just a little bit about science. I will attempt to dissect the paper and explain fine points that are not common knowledge to lay people. They are from Arizona, home to many a rattle snake. Rattlesnake crotoxin is a phospholipase A₂. It has other neurotoxic properties. There is a human non-pancreatic secreted phospholipase (nphsPLA₂) that makes an attractive drug target.

On to the paper

These authors started off their introduction citing literature showing a change in the lipid profiles of Coivid-19 patients. This group happens to study lipids, so it makes sense.

1. ↓phospholipids
2. ↑ Unesterified fatty acids
3. ↑ Lysophospholipids
4. ↑Acyl carnitines

All of this points to the smoking gun of phospholipase A₂.

The phospholipase A2 family of lipases has 12 family members that are secreted into the extracellular environment.  The sPLA₂-IIA has been associated with sepsis,

 Volcano plots are ways of displaying comparisons of big data sets.  The fold change between the two groups is plotted on the X-axis whereas the Log base 10 of, usually the p value, and in this case the false discovery rate (FDR) is plotted on the Y axis.  A value of -1 in log₁₀ is 0.1 or a false discovery rate of 1 in 10.  In these plots, the colored symbols indicate lipid biomarkers that go up or go down between the comparison groups. 

1A (&B) lysophosolipids are markers of Covid-19 severity

1C When enzymes have selective tastes in substrates

The first set of panels tells us that phosphatidyl ethanolamine (PE) is more likely to have its acyl group at the SN2 position in severe Covid than in normal subjects.  Both the mono and desaturated 18 carbon free fatty acids (FFA) are increased according to Covid severity.  Likewise, both acetyl and 2-hexenoylcarnitine are increased with Covid-19 severity. 

In the first set of panels of Figure 1C we see that our sPLA₂ has some picky tastes when it somes to substrate. Phosphatidyl choline (PC) and phosphatidylethanolamine (PE) are essentially the same head groups except that PC has three methyl groups (arrow) instead of hydrogens (not shown). One goes up in Covid infections, the other goes down. We an increase in 18 carbon fatty acids with one or two double bonds.

2. secreted PLA₂ as a biomarker of a bigger story

In figure 2A we see that the deceased Covid-19 patients had a significant increase in secreted PLA₂ in their blood compared to living groups. This panel contains some additional exciting information concerning population variations. Just because an enzymes is in the blood does not mean it is active. Panel 2B demonstrates that it is. Panel 2C demonstrates that the protein amount correlates with the enzyme activity. Panel 2D is a bit difficult to read. Go to the deceased row. Note that it is bluish on the top and deep read on the bottom. This is the same population distribution we see in Panel 2A. In those deceased patients with really high sPLA₂, were other Covid-19 disease markers elevated or decreased? Parse to the right and look at the other columns. O₂ saturation is pretty damn low in 2-3 of these high sPLA₂ red hot deceased patients. What about body temperature and the the role of PGE₂ in thermogenesis?

The publication does not end here. The authors go on to develop a plan to use secreted PLA₂ and other biomarkers to predict which Covid-19 patients will get sicker and which will be okay. Some other interesting ideas are presented that have nothing to do with PLA₂ in snake venom. Our secreted PLA₂ has a preference for phosphatidyl ethanolamine, an inner leaflet phospholipid. sPLA₂ would not see this phospholipid except in membrane fragments in dying cells. Extracellular vesicles with COX1 for prostaglandin synthesis were also mentioned.

In conclusion

There have been several misunderstandings that have created a social media $hit storm. Perhaps some things

1. We all use metaphors every day, often without regard that some may take our words literally.
2. The Bogna publication had a good mission to determine if toxin like peptides might exist in bodily fluids of Covid-19 patients. They may or may not have used reducing agents in their protocol. The reviewers may have missed this one. As interesting as this story is, they should have protein blasted their snake venom findings against he human geneome to determine if we make any similar proteins. They should also have searched their MS/MS fragments against the human database of proteins.
3. The Chilton Laboratory didn’t really do anything wrong in terms of press releases, in my opinion. Perhaps they could have done more to explain their findings and motives to lay audiences.