261007 – Pharmakeia, Venoms, and Exosomes Review

October 7, 2026 Leave a comment

Pharmakeia, Venoms, and Exosomes: A Fact-Checked Review

· @Thomas Lee Abshier, ND

Introduction

Sayer Ji’s essay “Pharmakeia: The Oldest Word for ‘Poisoned, Not Infected'” (October 3, 2026) is mostly well-sourced and careful about its own limits. The venom theories now circulating alongside it are not. This review checks both, claim by claim.

Ji argues that the Greek word phármakon meant remedy and poison at once, and that modern medicine split that single idea into “drug” (good, given) and “germ” (evil, invading). He builds the case on etymology, the history of virology, a 2018 mouse experiment on acetaminophen and exosomes, and the three uses of pharmakeia in the New Testament.

A second set of claims travels with the article in forwarded emails and social posts. These say that COVID-19 was snake-venom poisoning, that the vaccines carried genes for some thirty venoms, that gut E. coli keeps manufacturing them, and that common drugs such as ACE inhibitors, Ozempic, and remdesivir are made from venom. These claims trace mostly to Bryan Ardis’s 2022 film Watch the Water and to one Italian research group’s peptide studies.

The two sets deserve to be judged separately. Ji’s argument should not be discredited by the venom theories, and the venom theories should not borrow credibility from Ji’s footnotes. A naturopathic physician earns trust by holding his own side to the same standard of evidence he asks of the pharmaceutical industry.

The words: pharmakon, virus, poison, toxic

Verdict: accurate. Ji’s etymology is standard and checks out against the major lexicons.

  • Phármakon did mean drug, remedy, poison, and charm in one word, and pharmakeia carried the senses of drug-administration, poisoning, and sorcery. This is textbook classical philology.
  • Virus is Latin for poison or venomous sap. It entered English in the late 14th century as “poisonous substance,” and virulent means “full of poison.”
  • Poison does descend from Latin pōtiō, a drink or potion, and toxic from Greek tóxon, the bow, via toxikon, arrow-poison.
  • Beijerinck’s 1898 contagium vivum fluidum is the accepted turning point at which “virus” stopped meaning a poison and started meaning a filterable, self-reproducing agent.

Ji deserves credit for flagging his own weak points. He notes that iós (arrow) and iós (poison) are homonyms with different roots, a pun rather than a shared origin. He also reports that the link between pharmakós (scapegoat) and phármakon (drug) is called “probable, but problematic” by the scholar he cites.

The limit of the argument. Etymology shows what ancient people thought; it cannot show what is physically true. That “virus” once meant poison is a fact about language, not evidence that viral disease is really poisoning. Ji mostly uses it the right way, as a reminder of a forgotten unity of ideas, not as proof.

Benjamin Rush: the bogus quote and the real practice

Verdict: accurate, and to Ji’s credit. The well-known quotation in which Rush warns that medicine will “organize into an undercover dictatorship” is fabricated.

The University of Pennsylvania Libraries states that Rush never wrote or suggested that the Constitution should protect medical freedom alongside religious freedom. The library notes that versions of the false quote have appeared in print since 1899. The psychiatrist Thomas Szasz, no friend of medical authority, called it an obvious fabrication and pointed out that the real Rush went the other way: he treated an excess of the passion for liberty as a form of mental illness.

Ji’s description of Rush’s 1793 yellow-fever treatment by heavy bleeding and mercury purges is also historically sound. Opening an essay by correcting a quote that one’s own movement relies on is exactly the discipline the health-freedom movement needs more of. The rest of this review applies the same standard.

The 2018 acetaminophen exosome experiment

Verdict: the study is real and accurately described; the leap from it to “viral” illness is not supported.

The paper is Cho and colleagues, Scientific Reports 8:16070 (2018). Mice received a single 300 mg/kg injection of acetaminophen, which produced severe centrilobular liver necrosis. Exosomes harvested from their plasma were roughly 50–100 nm across and more numerous than in control mice. When given to cultured liver cells, they cut cell viability by about 20% and switched on JNK, caspase-3, and caspase-9. When injected into five healthy mice, they raised plasma reactive oxygen species, hepatic TNF-α and IL-1β, and hepatocyte apoptosis. Blood ALT did not change at four hours.

Ji’s summary matches the paper, and his own caveats are correct: it was an overdose model, the recipients’ ALT was unchanged, and healthy mice also make exosomes. The finding that injured cells send injury signals to neighboring cells is genuinely important. It also fits a long naturopathic emphasis on glutathione, since acetaminophen toxicity is driven by glutathione depletion.

Where the argument overreaches. Ji suggests that because these vesicles are virus-sized, a test that detects particles or genetic material “was not designed to tell the difference.” That is not how viral diagnosis works.

  • PCR tests detect a specific genetic sequence, not particle size. Exosomes from an injured liver carry the patient’s own RNA, not an influenza or coronavirus genome, and would not produce a positive result.
  • Viruses can be cultured from one host, passed to cells or animals, and shown to multiply. Exosomes do not replicate; they carry a fixed cargo and are used up.
  • Influenza-like illness spreads between people who took no drug. Fever and aches usually begin before anyone reaches for acetaminophen.

The defensible conclusion is narrower: drug toxicity can amplify and spread tissue injury inside one body, and that injury could be blamed on the illness the drug was treating. That is worth saying to every patient who doses acetaminophen heavily during a fever. It does not show that viral illness is mistaken drug poisoning.

The five-layer framework and the Revelation argument

Verdict: the scientific citations are sound; the framework is a useful lens, not a proven theory; the biblical reading is defensible as a pattern, as Ji himself frames it.

Ji’s “Poisoned, Not Infected” framework separates five layers that the word infection tends to lump together. Each rests on a real finding:

  • Initiating agent. Bacteriophages do carry the genes for diphtheria, cholera, and Shiga toxins. Calling pharmacology a branch of toxicology is a fair historical description; both study the effects of foreign substances on living tissue.
  • Receiving terrain. Melinda Beck’s work in the 1990s showed that a benign coxsackievirus mutated into a heart-damaging strain in selenium-deficient mice, and that the mutations were reproducible. This is one of the strongest pieces of evidence that host nutrition shapes viral virulence.
  • Accumulated exposome. Christopher Wild coined the term in 2005, and it is now a mainstream research field.
  • Mediating signals. Roughly 8% of the human genome is endogenous retroviral sequence, and the placental protein syncytin came from an ancient retrovirus. Both claims are accurate.
  • Interpreted story. Thomas Szasz’s Ceremonial Chemistry (1974) does open with the scapegoat theme.

The framework is strongest as a corrective. Medicine does tend to name one pathogen and stop asking about nutrition, toxic load, and the treatment itself. It is weakest where it implies the pathogen is incidental. Beck’s own experiment needed both a virus and a deficient host; neither alone produced the disease. Terrain and germ are partners in causation, not rivals.

One overstatement. Ji says he “debunked” the principle that the dose makes the poison. There are real exceptions: some endocrine disruptors act more strongly at low doses than at moderate ones, and timing of exposure matters during development. But dose remains the single strongest predictor of toxicity for most substances. “Dose is not the whole story” is accurate; “dose-response is a myth” is not.

Revelation. The noun pharmakeia appears in Galatians 5:20 and Revelation 18:23, with Revelation 9:21 reading pharmakōn (drugs) in some manuscripts, so “three times” is fair. Readers should know that the related words for sorcerers, pharmakoi and pharmakois, also appear in Revelation 21:8 and 22:15. The King James wording Ji quotes is correct. He states the scholarly counter-argument fairly and limits his claim to a recurring pattern—commercial empires that enchant and sell nations—rather than a prophecy about the modern drug industry. That restraint is appropriate.

The venom claims: toxin-like peptides in COVID patients

Verdict: a real but small, unreplicated study has been stretched far beyond what its own authors claim. The snake-venom theory of COVID-19 is false.

The study circulating with the article is Brogna and colleagues, F1000Research 10:550 (2021). Using mass spectrometry, the team reported peptide fragments resembling venom proteins—cone-snail conotoxins, snake phospholipases, metalloproteinases, and others—in samples from COVID-19 patients and not in controls. The “over 30 venoms” figure comes from its table of 36 matched proteins.

What the paper actually contains:

  • Very small samples. Plasma from 20 patients and 10 controls; urine from 2 patients and 2 controls; stool from 3 and 3. No age, sex, severity, or treatment data were recorded.
  • A tilted search. The fragments were matched against a reference set built from the complete catalog of venom proteins plus only a subset of ordinary proteins. A search weighted toward venom will tend to return venom matches.
  • Venom proteins have human relatives. Snake venoms evolved from ordinary body proteins. Humans make their own phospholipase A2, metalloproteinases, Kunitz-type inhibitors, and coagulation factor V—the same families on the list. Fact-checkers found that Ardis cited a rattlesnake-venom enzyme in COVID patients’ blood that is present in every healthy person.
  • No source found. The authors searched genetic databases and could not identify where the peptides came from. They speculated that gut bacteria might produce them in response to the virus.
  • Mixed peer review. One of the two reviewers answered “No” to whether the statistics were appropriate and whether the conclusions were supported.

Crucially, Brogna’s team never claimed that COVID-19 is snake venom, that venom was put in vaccines, or that anyone was poisoned on purpose. Their hypothesis is that the virus provokes the production of these peptides. The leap from their paper to deliberate envenomation was made by chiropractor Bryan Ardis in the 2022 film Watch the Water.

That film’s claims have been examined by PolitiFact, FactCheck.org, and Science Feedback, among others. They found that Ardis confused COVID monoclonal antibodies with antivenom, misread the studies he cited, and built part of his case on a plot line from the television show The Blacklist. A naturopathic reader should note that one need not trust these outlets in general to check the specific errors; each can be verified against the original papers.

What remains open. It is a fair research question whether COVID-19 triggers the release of toxin-like peptides, perhaps through the gut microbiome, and whether they contribute to neurological or long-COVID symptoms. Answering it requires larger studies, independent laboratories, and unbiased searches. Until then, the finding is a lead, not a conclusion.

Drugs derived from venom: ACE inhibitors, GLP-1 drugs, remdesivir

Verdict: two of these drugs have genuine venom ancestry, which is an honored story in pharmacology, not a scandal. None contains venom, and the harms attributed to them here are overstated or invented.

Venoms are refined by evolution to hit specific receptors with great precision. That is exactly why pharmacologists study them. Using a venom molecule as a design template is not the same as putting venom in a pill.

Claim What is true Verdict
All ACE inhibitors (“prils”) are made from Brazilian snake venom Captopril was designed from bradykinin-potentiating peptides first found in the venom of the Brazilian pit viper Bothrops jararaca by Sérgio Ferreira in 1965. Squibb chemists built a small synthetic molecule on that template, and the FDA approved it in 1981. Later prils were designed from captopril’s structure. All are synthesized chemically; none is extracted from snakes. Origin true; “made from venom” false
That is why ACE inhibitors destroy the kidneys ACE inhibitors are a standard treatment to protect kidneys in diabetic kidney disease. They can cause acute kidney injury in specific settings—narrowed renal arteries on both sides, dehydration, or combination with NSAIDs—and they can raise potassium. These are dose- and situation-dependent effects of blocking angiotensin, not venom effects. False as stated; real cautions exist
Ozempic is made from Gila monster venom Exenatide (Byetta, 2005) is a synthetic copy of exendin-4, a peptide first found in Gila monster saliva. Semaglutide (Ozempic) is different: it is built on the human GLP-1 hormone, with which it shares about 94% of its sequence. False for Ozempic; true for exenatide
Ozempic makes diabetes permanent, with damage lasting 20+ years after stopping Semaglutide was approved by the FDA in December 2017, so no one has stopped it 20 years ago. There is no evidence that it makes diabetes permanent. When people stop, blood sugar and weight often drift back up, which is the disease returning, not drug damage. Real concerns include nausea, gallbladder and pancreas problems, slowed stomach emptying, and loss of muscle along with fat. False
Remdesivir is made from venom Remdesivir is a synthetic nucleotide analog designed by Gilead to jam viral RNA copying. Gilead called the venom claim entirely false, and Ardis’s main evidence was that both liquids have a similar yellowish tint. Its benefit was genuinely disputed: the WHO issued a conditional recommendation against it for hospitalized patients in November 2020. False; efficacy concerns legitimate
The COVID spike protein is derived from snake venom A 2020 PNAS paper by Cheng and colleagues noted short stretches of the spike that resemble motifs in cobra and krait neurotoxins. Short similarities of a few amino acids occur by chance across unrelated proteins. The spike is a coronavirus protein whose relatives exist in bat coronaviruses. False
Vaccines contain venoms Ingredient lists for licensed vaccines are public and contain no venom. No analysis has found venom proteins in them. False

The fair point underneath. Every one of these drugs has real side effects, and patients deserve frank disclosure of them. That case is stronger when it rests on documented adverse effects than on invented origins.

Manufacturing claims: E. coli, nanoparticles, gut replication, water supplies

Verdict: one starting fact is true, and one real manufacturing controversy exists. The chain of claims built on them is not physically possible as described.

“The vaccines were cultured in feces.” The mRNA vaccines do begin in E. coli. A circular DNA template, called a plasmid, is multiplied in laboratory E. coli, purified, and cut open. The mRNA is then copied from that template in a cell-free reaction, and the DNA is broken down and removed. Laboratory E. coli K-12 strains are domesticated bacteria grown in sterile broth, not fecal material. The same method has produced human insulin since 1982.

The real controversy: leftover DNA. In 2023 independent researchers, including Kevin McKernan, reported fragments of the plasmid DNA remaining in vaccine vials, including a short SV40 promoter-enhancer sequence in the Pfizer product. Health Canada confirmed that Pfizer had submitted the full plasmid sequence without flagging the SV40 element, and stated that the fragment is not the SV40 virus, is inactive, and is below regulatory limits. Whether those limits are adequate for DNA packaged inside lipid nanoparticles is a legitimate scientific question. It is also a question about DNA, not venom.

“The LNPs carried the genetic material for 30-plus venoms.” False. The vaccine mRNA sequences have been published and sequenced independently, including by critics of the vaccines. They encode the spike protein, not venom proteins.

“Gut E. coli keeps replicating the venoms.” This cannot happen as described, for three reasons:

  1. Bacteria cannot copy mRNA. They have no enzyme that makes RNA from an RNA template, so injected mRNA cannot multiply in them.
  2. Vaccine mRNA is built for human ribosomes, with a human-type cap and tail. Bacterial ribosomes need a different start signal and translate such mRNA poorly if at all.
  3. Injected mRNA stays mostly near the injection site and lymph nodes, where it has been detected for up to several weeks. There is no known route by which it reaches the colon intact in meaningful amounts.

Brogna’s group did report, in a separate preprint, toxin-like peptides from bacterial cultures exposed to SARS-CoV-2 and speculated about gut bacteria. That is an unconfirmed hypothesis about the virus and the microbiome. It says nothing about vaccine mRNA.

“LNPs could have seeded municipal water.” There is no evidence for this. Lipid nanoparticles are unstable outside cold storage, would be diluted by billions of gallons, and would be degraded by chlorination. Swallowed mRNA is destroyed by stomach acid and digestive enzymes. This idea comes directly from Ardis’s Watch the Water, and no water test has ever supported it.

“LNPs were designed to look like exosomes.” Partly true in spirit. Both are tiny fat-walled bubbles, and drug-delivery scientists openly borrow ideas from natural vesicles. Resemblance in size and wall material does not make LNPs viruses or carriers of venom genes.

Exosomes versus viruses: can the particles be told apart?

Verdict: the size overlap is real and has caused genuine mistakes in tissue photographs. It does not follow that the viruses seen by electron microscopy are mostly exosomes.

The idea circulating with the article is that cells make exosomes to carry poisons out of the body, and that what microscopists call viruses are really these exosomes. The first half has support. Cells do use extracellular vesicles to export waste, misfolded proteins, and even drugs; cancer cells use them to expel chemotherapy agents. The 2018 acetaminophen study also showed vesicles carrying injury signals.

The second half fails on several independent tests:

  • Genome. A virus carries its own genome, which codes for proteins the host does not make. SARS-CoV-2 has been sequenced millions of times, and its genome is not found in uninfected human DNA. Exosomes carry fragments of the host’s own RNA.
  • Structure. Cryo-electron tomography of intact SARS-CoV-2 shows regular particles studded with a characteristic number of spike trimers and a packed ribonucleoprotein core. Exosomes are irregular bubbles with host surface markers such as CD63 and TSG101.
  • Replication. A single virus particle placed on susceptible cells produces many thousands of copies of itself, which can be counted by plaque assay and passed on again. Exosomes cannot reproduce.
  • Transmission. Viral illness spreads from person to person, and the same genome is found in each link of the chain. Exosomes do not travel between people.

Where the concern has merit. In 2020, CDC electron microscopists including Cynthia Goldsmith warned in The Lancet that several published autopsy photographs labeled as coronavirus particles were actually normal cell structures, such as coated vesicles. That is a real error, and it shows that a photograph of a round particle inside tissue proves little by itself. The correction, however, came from virologists who could tell the difference, which shows the difference can be told.

Researchers who study the overlap seriously, such as those behind the “Trojan exosome” hypothesis for retroviruses, treat viruses as real agents that borrow the cell’s exosome machinery. Ji’s own article takes that position. The stronger form of the claim—that viruses as a whole are misread exosomes—goes beyond what even Ji argues.

Summary of verdicts and conclusion

Ji’s essay largely holds up; the venom theories attached to it do not.

Claim Source Verdict
Phármakon meant both remedy and poison; virus meant poison Ji Accurate
The Benjamin Rush “medical dictatorship” quote is fake Ji Accurate
Acetaminophen-injured liver cells release vesicles that injure healthy cells Ji Accurate (overdose, in mice)
Such vesicles could be mistaken for viral illness by testing Ji Not supported
Host nutrition and toxic load shape how disease unfolds Ji Well supported
“The dose makes the poison” is a myth Ji Overstated
Pharmakeia in Revelation describes a recurring imperial pattern Ji Defensible interpretation
Toxin-like peptides were found in COVID patients Brogna et al. Reported once, small and unreplicated
COVID-19 is snake-venom poisoning Ardis False
Vaccines carry genes for 30+ venoms Circulating False
Gut E. coli keeps making the venoms Circulating Not biologically possible as described
Venom genetic material was put in water supplies Ardis No evidence
Viruses seen by microscope are mostly exosomes Circulating False, though some tissue images were misread
ACE inhibitors are venom and destroy kidneys Circulating Origin true; claim false
Ozempic is Gila monster venom and makes diabetes permanent Circulating False
Remdesivir and vaccines contain venom Ardis False
Residual plasmid DNA, including an SV40 fragment, is in the Pfizer vaccine McKernan; Health Canada True; health significance debated
Americans face heavy, poorly studied chemical exposures Circulating Supported
The poisoning is deliberate Circulating No evidence

Conclusion. The best part of Ji’s argument is a call for humility. Every remedy is also a potential poison, the body’s response to injury can look like infection, and medicine too often looks for a single invader while ignoring nutrition, toxic load, and the treatment itself. A naturopathic physician can affirm all of that without reservation.

The venom theories work against that message. They take one small, honest study and a fact of drug history, then add claims that contradict basic molecular biology and the drugs’ own published chemistry. When such claims are shared alongside sound criticism, they hand opponents an easy way to dismiss the sound criticism too.

The health-freedom movement is most persuasive when it practices what Ji did in his first paragraph: correcting its own errors before anyone else has to. Honest skepticism of the pharmaceutical industry and honest skepticism of our own sources are the same discipline.

This article is for education and is not medical advice. Do not stop a prescribed medication without speaking to the clinician who prescribed it.

Sources

Leave a Reply

Your email address will not be published. Required fields are marked *