An interesting paper was forwarded to me this week by a colleague that would prefer to remain anon at this point. I initially didn’t think much of it until I excavated the entire paper from the paywall and read the thorough work performed on these short RNA sequences. It’s a very thorough characterization of these localization signals. This lead me to three other papers that are worth a read. They hopefully give you a sense of how complicated RNA biology is and how nascent our understanding of the role of PseudoU is in cell regulation.
TL/DR: Certain RNA sequences are zip codes for RNA localization in the cell. One known as SIRLOIN (SINE-derived nuclear RNA localization) or CCTCCC is common in RNAs that localized to the nucleus. Its a short 6bp sequence you would expect to see every 4096 bases (4^6) and I can’t find any in SARs-CoV-2 but I was alerted to one being found in the 3’UTR of the BNT162b2. I then searched Moderna and C19 sequences and found 5 of them in the mRNA-1273 and none in C19. If you losen the search up to just the 5mer (CCTCC) you begin to find more in Modern and Pfizer and some in close proximity to each other. This proximity likely matters according to Lubelsky et al. The codon optimization used for these vaccines seems to have introduced these sequences to the mRNAs.
I’m not certain what this means. Nuclear localization is probably not helpful for spike expression but it is also not a good sign for reverse transcriptase activity that could lead to gDNA integration of vaccine derived mRNAs. This is obviously a very hot topic after Alden et al. and Zhang et al. published work suggesting mRNA vaccines might reverse transcribe and potentially integrate to the human genome.
One should keep in mind that CCTCCC is actually CCUCCC in RNA and in modRNA terms it is CC-N1methyl-PseudoU-CCC. It’s not clear what the N1PU would do to this recognition sequence but the PUS enzymes are nuclear localized as are most natural non coding RNAs that are found to be containing Pseudo uracil. If I had to guess, N1-methyl-Pseudo U is more likely to help than harm the nuclear localization.
The techniques described in the above papers could easily answer this question. I hope someone has the courage to test this.
Let’s have a look at these sequences. Note the pattern matches in boxes.
What makes these sequences more compelling is when you find CCTCC in close proximity (40-50bp) of another CCTCC as seen in Lubelsky et al.
As seen in Lubelsky et al-
This isn’t a perfect hit but it should certainly raise some questions regarding what has been done to monitor localization of the mRNAs injected into a billion people without liability.
Anyone wanting to obtain research papers, but lacking the substantial subscription fees or access to institutions with subscriptions can usually find these papers at Sci-Hub. Just copy the DOI number beginning with "10." and paste into the sci-hub search bar.
Thanks for looking into the details of the mRNA technology used in the vaccines!