Mutations such as deletion of a gene or gene region don't revert very easy. Single point mutations can easily revert.

Are you sure that you produced the virus with the mutation? Because how can you sequenced your virus at passage 0 and know it worked?( if you do that you probably sequenced your infectious clone). Did you analyzed all the viral populations when you sequenced it at p1? Your reverted virus is the consensus sequence i guess, but did you have minor variant harboring your mutation? Is the virus able to replicate correctly with the mutation? maybe at the first round of replication RNA polymerase produced the virus with the reversion and it emerged because it has an higher replicative fitness....

Dear Franck Touret ,

We did sequencing on our newly developed mutant (P0) to check the correct mutation has been achieved. The mutant virus is viable and we managed to passage it up to P3, and we sequenced each passage to see if there is a change on our mutant

Ok, so it replicates correctly until p3, not p1?which method did you used to sequenced it at p0 and how do you discriminate it from DNA/RNA resulting from your transfection? Do you have all viral populations at each passage to know when you revertion occured? If it goes to p3 can’t you use your p2 for your experiments? Also Maybe this is a specific cell line adaptation? Did you try mammalian cells like veroE6 or HUH 7.5?

In order to minimize mutations during Passaging you might infect with a high MOI this will result in less Virus generations and less potential mutations. In case your mutations aims to create an amino acid substitution aboid single nucleotide exchanges try to exchange 2 nucleotides per triplet. This reduces the potential of reversions to a high degree.

Franck Touret actually the reversion occurred as early as P1 and we used Sanger method to sequence it. We also suspect that this is specific cell line adaptation/fitness and we planned to passage it in mammalian cell lines. But just in case, the same thing occurs, we need a back up plan to stabilize it

Jan Baumann thank you for your suggestions

Not knowing exactly what your virus mutation is, I can't do a direct estimation. However, dengue is an RNA virus. That means it should be expected to have a mutation rate of about 10^-5. (This varies somewhat.) Dengue is ~10700+ nt. https://www.ncbi.nlm.nih.gov/genomes/GenomesGroup.cgi?taxid=12637

That means that per copy event, my starting estimate would be a mutation rate of ~2 per copy event. That will give me a guess for the analysis method used in Drake/Holland . In an infection event, there will be a large number of copies made in each cell. Drake shows that the copy system used by most RNA viruses means that you can treat each cell infected as a single copy event. (Although after the copy cycle is complete, the next cells infected are another event.) At each iteration there is selection. Article Mutation rates among RNA viruses

What your lab culturing experience is telling you is that the mutations that you made are insufficient to prevent reversion. (Yes, I know you knew that.) I do not believe based on your passaging experience that you are likely to be able to make this vaccine safe.

I would suggest that one way to deal with this is by removing a major gene that still allows replication. One method would be removal of an immune system modulator. However, raising antibodies to the immune system modulator would be a major component of a successful sterilizing vaccine.

Another method would be to create a chimeric system with a DNA virus for stability, and express the critical dengue genes there in multiple virus cultures. Then the vaccine could be made by bringing those together. One could do that with Adenovirus, or with AAV, for instance.

For myself, I would lean toward the latter, or even go to an entirely synthetic vaccine with no live components. This last is because the challenge with dengue is eliminating original antigenic sin. To do that, I think you will need to link the virus to highly immunogenic presentation molecules to get a satisfactory result for all 4 primary strains.