Two identical point mutations occured after sequencing of two of my colonies (DH5alphas+pGEMTeasy with insert).
To amplify the region of interest I have performed a genomic PCR with Pfu polymerase, adding 10% glycerol to the mix, since it took me a lot of attempts to make the reaction work properly. I never used more than 30 cycles, and annealing temperature was 61 degrees. I isolated from gel (sybrsafe gel) and re-amplified my product with primers with restriction sites, again with Pfu. Then I ligated to pGEMTeasy and transformed DH5alphas.
What could possibly have happened?
Since the mutations are exactly the same, it is probably the PCR failing on the first few cycles, but I don't know how likely that is, since my product is only 1.5 kb long and Pfu is a high fidelity polymerase. Could it be the glycerol? I didn't find anything suggesting that it might induce mutations though...
Could the fact that it is a GC rich region I'm trying to clone contribute to this somehow?
Suggestions will be very much appreciated!
Yes it could be entirely coincidental that the mutation occurred during an early round of PCR amplification. You could test this by repeating the amplification and clone and sequence a few. It could also be that your source strain has picked up the mutation so all of your clones will carry this same allele.
Point substitutions are not typical PCR errors. In G/C rich regions, you would normally see larger rearrangement, skipping, or it not working at all.
More likely I'd suspect a sequencing error. High G/C can absolutely cause this in a systematic way. Look at the chromatograms / peak shape. Do they look in any way off? And sequence in the other direction.
Second it might be that your gene is actually different from what you think it its. It could be database error, or simply natural variation.
Hi there...
A single point mutation may be a result of sequencing error. You need to check the chromatogram. If the all the sequences are alright except the mutated one, then there must be some error during the PCR cycle or the database error.
Thank you all for your answers!
@Michael: my current approach is indeed trying to re-amplify and use several PCR products to use as inserts, in order to sequence them and see if these mutations occur again. I wish I could re-test my previous ligation mix, but I don't have any left.
@John & Gaurav: this is really interesting. I used several sequencing reactions, so that the resulting sequences would overlap and I could compare them in case I had bad peaks, half of them were forward reactions, and half reverse. Indeed one of the reverse reactions failed to work in both colonies (exactly in the same region, actually containing the point mutations) and gave back a completely worthless chromatogram; the overlapping forward reaction worked perfectly instead. However, I could not compare the two sequences in any way. Could this be caused by the high percentage of GC?
In any case, the peaks look very neat. I'll attach a screenshot of both point mutations (highlighted).
The mutations are transitions (G->A and C->T). I assume that makes it more likely for them to be sequencing errors?
And finally, I'm sure the assembly is correct, because it has been used before and the sequence appears to be exactly the same.
Certainly does not look like sequencing error. I think you can assume it likely is an errant PCR reaction, but you will know that with your next round of cloning, OR you have a error in the template, if you get the same mutations next time, that is probably the situation. You could also confirm by just directly sequencing the PCR product itself before cloning.
Also growing heat-shocked transformed cells for more than an hour before plating allows for cell division to occur and gives rise to multiple colonies from a single ligated DNA molecule.
In my experience, point mutations are the only type of mutations I've seen by PCR (never rearrangements, deletions, etc.), with transitions the most frequent. Point mutations are quite common with regular Taq but I have not had a single one with proof-reading enzyme (NEB Q5).
I agree with Michael, and suspect the template. Can you sequence that?
- Badges
- Science topic