Why is the new mRNA vaccine unable to modify human DNA?

Question

One of the most common worries expressed against the new coronavirus vaccine is that, supposedly, injecting mRNA into a cell has the potential to alter the human genome.

I am constantly bombarded by my family with links to conspiracy theory and vaccine hesitancy websites. Notably, these links often feature scientists with high academic degrees.

One example of such links features professor Roman Zieliński, who -- among other claims -- warns that the new vaccine has the potential to alter human genome. Notably, his field of specialization is genetics.

The explanation he gives looks convincing and sound, at least to a layman like me. I will try my best to translate the parts of the interview I linked to that are relevant to this issue:

Journalist: Claims are made that mRNA included in the 'vaccine' against the SARS-CoV-2 virus (commonly called C-19) may have an impact on our human DNA. In what way? And what does it mean for an average Joe?

Roman Zieliński: Your question brings up Watson and Crick's Central Dogma of Molecular Biology. These scientists had earlier received the Nobel prize for developing the double helix DNA model. Accordig to their Dogma, which is a part of the foundations of molecular biology, the genetic information does not only flow from DNA through RNA to protein, but also in the other direction, from RNA to DNA. Information, therefore, circulates between these two nucleic acids. This means that every RNA molecule that enters a cell may be transformed into DNA and may join the genome.

And this indeed happens. Viral retrotransposons incorporated into the human genome are one example. Around 60% of the human genome consists of them. tRNA and mRNA move towards the nucleus in a similar way. Cells have enzymes known as reverse transcriptases that enable rewriting RNA into DNA. Such enzymes include the human telomerase and reverse transcriptase of viral origin.

The process of reverse transcription, that is rewriting ribonucleic acid into deoxyribonucleic acid inside a cell is an evolutionarily old mechanism. It would be odd if mRNA molecules brought into our cells by this "vaccine" was exempt from this mechanism. There are numerous threats posed by introducing this mRNA construct in this "vaccine". The possibility that it includes itself into the human genome and further impacts the expression of genes and their mutations is one of such threats. The potential of transgenes to impact genome cannot be overestimated and the results of genetic research on plant GMOs support this. Their mutagenic potential on the genome is even stronger than the potential of chemical mutagens.

I tried to do my research. I entered the terms Zieliński used into Wikipedia. It seems that the central dogma of molecular biology is indeed an instrumental principle of this field, even though it is called in a suspicious way ("dogma") and this principle indeed explicitly states that information may be transcribed from RNA into DNA. The Wikipedia article about telomerase indeed states that this is an reverse transcriptase enzyme present in human adults in, for example, "in male sperm cells (...) in epidermal cells, in activated T cell and B cell lymphocytes, as well as in certain adult stem cells". And so on.

I found a relevant Skeptics.SE answer: Can an RNA vaccine change your DNA permanently?. It states: "mRNA produced in a cell doesn't modify your DNA, and neither would an mRNA vaccine.", but this seems to contradict the aforementioned central dogma of molecular biology.

Finally, and perhaps most importantly, I found mainstream scientists deny such claims:

• "Injecting RNA into a person doesn't do anything to the DNA of a human cell," says Prof Jeffrey Almond of Oxford University.;
• Prof. Krzysztof Pyrć, responding directly to similar claims made by Zieliński on another occasion, says he cannot even imagine how can mRNA work in that way (impact the expression of genes).
• Dr Maciej Przybylski says a human body cell does not contain enzymes necessary for RNA to be included in human genome _{(but Zieliński mentioned such enzymes!)}

Unfortunately, these are just denials. These scientists repeat that mRNA included in the vaccine cannot impact human DNA, but do not elaborate why and how it cannot.

This is, therefore, what I would like to ask in this question. Why is it not possible for the mRNA included in the vaccine to impact human genome? In particular, how does the central dogma of molecular biology not mandate such a possibility?

Answer 1

Roman Zieliński seems to be intentionally misleading you by making an implausible circumstance that is technically possible sound like a likely outcome. This strategy is not unusual among people who argue against vaccination, because they have very little actual science to argue based on.

The Central Dogma as stated by Crick says nothing about RNA vs DNA. It does not say that RNA must enter DNA, or anything like that. It only says that information in protein sequence does not become information in nucleic acid sequence, and that information in nucleic acid sequence does become information in protein sequence. The historical basis of this is that there was a time in which people thought protein might be the hereditary material. We know now this is not the case, and Crick was trying to make a special point of it by giving it this "central dogma" name.

Telomerase is a special reverse transcriptase that adds a "cap" to human chromosomes by replicating a specific RNA it holds. It does not copy random sequences of RNA to DNA or put DNA anywhere else.

Reverse transcriptase of viral origin comes from retroviruses. You don't have these reverse transcriptases present unless you are infected with such a virus, like HIV. These viruses do not insert all RNA into the genome, they use another protein called integrase that specifically inserts virus-derived double-stranded DNA copies into the genome. Even if this process were less specific, the only affected cells would be those already infected with HIV.

Similarly, he might be making reference to endogenous retroviruses but these do not incorporate random RNA into the genome as he implies.

If random RNA present in cells regularly inserted itself into the genome, we would have a huge huge problem all the time, not just when adding some exogenous RNA. Human cells are full of all sorts of mRNA, and if it was inserted back into the DNA genome we'd have multiple extra copies of all sorts of genes. They'd be in the wrong places, getting transcribed under the wrong promoters, ending up in the middle of other genes, etc.

Answer 2

To supplement Bryan's answer (although I'm not going to say anything fundamentally different here), according to their proponents, mRNA vaccines are considered the safest genetic vaccines (in this integration regard) because mRNA transcription ("transfection") to proteins happens outside the nucleus. Quoting from a review

mRNA vaccines do not interact with the host-cell DNA, they avoid the potential risk of genomic integration posed by DNA-based vaccines.

On other hand, if you're really curious about this, DNA vaccines have also been proposed and even tried in clinical studies, and regulators are a bit more concerned about integration effects with these, even just at the local, administration-site level.

employing DNA as a basis for vaccination also implicates some disadvantages. A concern in this context is the long-term persistence of DNA plasmids upon injection. Indeed, DNA persistence was shown in various preclinical studies that demonstrated the presence of plasmid DNA for up to 2 years upon IM injection with low but detectable expression and immunogenicity in a mouse model. According to the FDA, DNA persistence is not generally evident at ectopic sites in biodistribution and persistence studies, but remains detectable at the injection sites for periods exceeding 60 days. Especially in the context of this long-term persistence, the presence of foreign genetic information in the nucleus of transfected cells poses the additional risk of genomic integration into the host's chromosomes and the resulting threat of mutagenesis and oncogenesis. Despite negative results in several studies focusing on detection of DNA integration events upon IM injection in small animal models, genomic integration events were detectable following electroporation in mice demonstrating that integration represents a small risk that nevertheless needs to be considered in systems with enhanced DNA uptake. The FDA recommends integration studies to be included whenever plasmid DNA exceeding 30,000 copies per μg of host DNA persists in any tissue by study termination. The WHO advises integration studies as part of the preclinical safety program of DNA vaccines.

If you read that carefully, the regulators are concerned about DNA vaccines giving the patient some form of cancer, and so they require tests showing low integration effects, however this [integration] is generally regarded as a remote risk. Note that this kind of concern didn't stop DNA vaccine from proceeding to clinical trials, e.g.

DNA based vaccines were among the first to proceed to clinical trials upon the Zika crisis in 2016.

In general, DNA vaccines have proven to be less effective in clinical trials in part because they need to cross two membranes (cell and then nucleus) to actually have their desired transfection effect. mRNA vaccines have required quite a bit of engineering just to achieve the former (i.e. cell entry) well enough--the whole liquid nanoparticles thing is about that.

vaccination with a DNA vector alone generally leads to relatively low immunogenicity, especially in large animal models and humans. A factor that may play a role is the need for DNA vaccines to cross two cellular membranes, i.e., the plasma, as well as the nuclear membrane, in order to achieve protein expression. Of note, this does not hold true for RNA vaccines, which are translated upon crossing the plasma or endosomal membrane, respectively.

On the other hand, DNA vaccine proponents point out that even for these the risk of integration has been overblown, and that even DNA vaccines are not considered gene therapy (i.e. host-DNA modification):

DNA vaccines did not need to be evaluated by the US National Institutes of Health (NIH) Recombinant Advisory Committee prior to human clinical trials, unlike viral vectors for gene therapy. Nevertheless, significant safety studies were initially required to evaluate the possibility of integration of the plasmid DNA into the host genome. As a result of these studies for both human vaccines and for the licensed DNA vaccines for fish, as well as the many human studies with DNA vaccines that have demonstrated safety, little concern now exists regarding integration. Comparisons have stated that mRNA offers an advantage because RNA itself cannot integrate into genomic DNA without the presence of the viral elements in a retrovirus that enable such integration (reverse transcriptase and integrase). However, HERVs (human endogenous retroviruses) whose remnants are now permanent parts of human genomes as retrovirus-like sequences comprise up to 8% of the human genome. In addition, some recipients of mRNA drugs or vaccines may be already infected with a retrovirus (e.g., HIV), thus providing a theoretical means for provision of the proteins needed for integration. Nevertheless, the risk of integration remains, at this point, extremely unlikely for mRNA, even from a theoretical standpoint, nor is it any longer a significant concern for plasmid DNA. This means that mRNA does not offer any clear advantage compared to plasmid DNA in this regard. From a regulatory perspective, mRNA prophylactic vaccines appear to not be considered gene therapy products, similar to DNA vaccines before them.

In other words (as Bryan said and you can see reflected in the quote above), the only (theoretically) known means by which such integration of mRNA into cell-host DNA could happen is co-infection with a retrovirus, which is regarded as an unlikely coincidence/risk. There don't seem to be any experiments that have actually tried to even demonstrate this risk, i.e. that mRNA (particularly one from a vaccine) could be integrated in the host in the presence of e.g. HIV.