My prior analysis of Pfizer had (mistakenly) used an very large antisense open reading frame that encodes a protein unknown to science.

In my article, Will Omicron Save Us from Vaccine Induced ADE?, I incorrectly asserted that the Pfizer mRNA encoding spike protein did not contain the ADE-related epitope, LYQDVNC, and I speculated that Pfizer left it out to avoid disease enhancement.

A couple of smart readers double-checked and found the ADE-related epitope, LYQDVNC (thank you, runaway and gibarian1979 - see their comments on the original article.)

Retracing my steps in the analysis, I found that in my haste I had grabbed the right open reading frame (ORF) from Moderna, but the wrong ORF from Pfizer. It’s a rookie mistake, but you’ll see in a minute why I made it.

Wikipedia: Open Reading Frame

The bottom line is yes, Pfizer’s Spike-protein encoding open reading frame also encodes LYQDVNC, the disease-enhancement epitope discussed at length by Dr. Fantini and colleagues. In fact, in spite of different codon usage, Pfizer’s spike protein matches Moderna’s across the length of the mRNA ORF1 (this fact is not news to anyone).

Why I Made The Rookie Mistake - But Look What I Found As a Result

So, the reason why I grabbed the wrong ORF was because it was so large! (I had already analyzed the Moderna sequence and “knew” that the longest ORF was the spike ORF. And the ORF webapp places the longest ORF at the top, so… Like I said, rookie mistake.

Here’s the weird wrong ORF I grabbed (ORF20). It’s encoded by the anti-sense of the Pfizer mRNA. It’s 1,295 amino acids long, uninterrupted - and it’s longer than the Spike protein (1,273 aa). Which is, well, weird. In decades of bioinformatics analysis of mRNA-encoded proteins, I never saw an antisense sequence that also encoded a protein over the entire length of the sequence I was studying. (More discussion below)

>lcl|ORF20

MQYQLESSCVVQFHALQHGLRIVLVELAAAATATTALQAATAAGHATQHD

CDHHDGNQSGDKAQPDVPGPLDVLLVLPQFLQVDQALVQILGHLVQPVDL

FLDVHDAGIDSADIAQVHVGACVVLKVLVQFLFEAVQLGLQRVVHGIVHN

ADHDVAVARHEGVVGGDDLGLVEVPLCHEPMGAVGHEHAFSRKVGFAVVA

DGWSGGEILLLSGHICHVQKHHAVRGRLREAHQVVALAAKVHSLALAQHT

LRHLGGGQIGRGSNLGGSDQLLGHVCLEALQSACDQSVDLHLGLRRVQSA

QDIVQHRADGAEVGGQLLDQGVQCLGILVDHVLQLSQGACCAAQAVLDLA

DGAVELVGDQLLVLVQHILGHSDAVEPVGHLHSKGDLQSGACSKCPAACD

CAGQQGRCVLGDHLIGQQRRQHCQSVKLLGANQIPGGNVAQTIAILLDEA

GVGQCHFVEQQVLDEAPLAGLARIGQNLAEIEAAEVLDRRGLVDLLHLGE

HLLGVLVLFHGDPCQGSIQLGAEAAVLQQQVGALGGIAADVHGAVHAGLG

HGHRQDLCGHADGEVGGDSDRVVGVGHAVLGAQRHCVGNDALAGHASGSP

RAVCLCLVAGADSSADGDVALVAIVHVLGSDQTAGSGLKHIAAGGVHPPC

RCQLIGVNGHGHFGTVHVLVQHCHLIAGVGARGDHRHSAEAARGDVQDFQ

CLGISNGVCGIGDIPAKLLEWQELLVALCQHAGAGQAVEVEVHAFVLHEI

GAFLRAAHCGRGMQQFEAQHHHSVGLIAHAICGLKAVGLQWEVAVEAFHA

VTRGAAGLIDLGGDVPLEGLQIGLPEQPVQVIVVAADFGVQVVAVPGNHT

AGEVVGQLVVVVGDLACLSRGNLPHFISPDHEAVGVHVCEAQVVQLGRGH

AVALEGAEAGGVVQHGVVGHAIADPLPVPGVHRGESGGIEHLAEGAQIGD

IGEPHDGFGGLHPEVAGLVDALFHGEGLQGALCLAQRIQSTIHGVGDGAV

LVVLQQEGSRLQVAHIVSGGTSCPSAAAIARCQVASVQGQQCLKPGDVDA

DGQIHQGFQSREALRQIPHEVDRGVLAVDLEVAVDVLKHELAQVLEVALL

AFQVHQERLGHVLEGAVVGAAVHPELAFHPALVVLVVVDAQEGVVAELEL

AHFDDHVGGVVHDQQALGLAVQCGAEDPASDDVGLLGAGKVHPVVEGQHG

VVESLGAIGAGHVDGVEPGHVAEERQEQVLGRVQHAGSEHLVGVVHASGK

AVGVGWRQLCSGGQVHTLAGHQRQQHQEHEHGGGFSLSLWGPEEY

You almost never see a long ORF like this in the open reading frame analysis unless it’s a functional biological entity. There are a few pairs of genes in the human genome that have another gene also encoded by the anti-sense strand (they are known as SAS (Sense/Anti-sense gene pairs), but they are rare (see a paper on this from 2013).

However, a protein blast analysis shows that this amino acid sequence has no homology to any known protein in the entire NCBI protein database.

I checked the Moderna ORFs again, and, oddly, weird thing - it’s not there at all. There is no ORF20 in Moderna’s mRNA.

Somehow, Pfizer chose codons for their spike protein that encodes the spike that is 100% identical to the Moderna spike protein (via identity to SAR-CoV-2 Wuhan sequence), and also managed to encode an extremely long additional protein encoded by the antisense strand. Meaning somehow they never encoded a stop codon - and there are three chances to do that - UAG, UAA, UGA.

There is this study , which reports that Pfizer changed their nucleic acid sequence for the Spike protein to optimize codon usage. So evidently a lot of thought went into codon selection to get more spike out of each cell for the buck.

What This Could Mean for the Pfizer-Vaccinated

There have been reports that the mRNA from vaccines can be reverse-transcribed into human cells; that means that this strange large protein could also be written into the human genome (in some of the cells) of some people receiving the Pfizer vaccine. The conditions are that the cells have reverse transcriptase available, or that a dormant virus becomes activated that encodes an intracellular reverse transcriptase.

I ran protein structure prediction modeling using Phyre2 on the Pfizer ORF20 antisense protein.

Here’s the predicted structure:

According to Phyre2, the structure maps to pandonodin, specifically, lasso peptide pandonodin. The result comes with a warning:

Confidence in the model: 28.6%

WARNING: *Very* low confidence model. Please treat this model as highly speculative

Nevertheless, there may be unanticipated biological activity if this protein were to be expressed by human cells. The likely effect? An immune reaction against a foreign protein. Was Pfizer shooting for adjuvantive effect? Who knows. Dumb luck? Maybe. Pandonodins are also being explored as “reporter systems” given their ability to fuse with other proteins. Speculative indeed.

We know that mRNA biologics are being injected into people who are also infected with other, reverse transcribing viruses. Perhaps those designing mRNA biologics should be required to encode some stop codons in the anti-sense strand of their biologics just to be sure they don’t introduce a new pathogenic protein.

More info on lasso peptide pandonodin for the curious:

https://pubs.rsc.org/en/content/articlehtml/2021/sc/d1sc02695j

https://pubs.acs.org/doi/abs/10.1021/acschembio.9b00676