Hi all, it might be a very basic question, but I need some opinions:
I need to separate a 294 bp wild type PCR product from a 232 bp knock-out PCR product. I want to do DNA extraction out of the gel and then sequence the bands. The lower band worked quite well last time, but the sequences of all the upper bands were very messy. I was thinking that maybe the products only separated insufficiently. So now I try to adapt the process. So far, I used a 2% agarose gel, 120 V, around 45 minutes.
What would you recommend to change and how? Also, do you think, the incomplete separation could be the root of the problem or might there be something else?
Thanks!
I agree with Chris Cromwell
that we need to see a picture of the gel and ideally the .abi file of the sequence of the upper band. It is possible that the upper band is a mixture of large/large and large small heteroduplex if the 2 bands were generated from the same dna sample and this would not sequence well. Possibly also the large band is GC rich and has secondary structure sequencing difficulties. I agree that the gel concentration is ok but running for longer and also at a lower voltage to minimise heating and thermal diffusion of the bands is a good idea. If all you want is good diagnostic separation then digesting with a restriction enzyme that only exists in one amplimer will increase the percentage size difference and so give better separation of the amplimers while still giving the existence of 2 amplimers
Hello all, thank you for your answers! I attached photos from the gel for both my targets and also two representative .ab1 files.
When redoing the process last week, I used a longer 3% gel, 90V, 3 hours in total (I checked in between, but the bands from target1 still didn't separate too well). The bands became faint but I still hope for a better resolution this time. I will keep you updated and will be thankful for any further comments, thanks for the input. Chris Cromwell
Paul Rutland Foad AlzoughoolEDIT: I have two targets, one of which is the previously described 294 bp product, the other one is a 501 bp product. I described the first target, because I experience the most issues there, but also with target2 I get messy sequences for the upper bands, even though the difference in size is much bigger from wt to knock-out here. But, in essence, it's the same problem, so I hope this doesn't confuse.
Thank you for the files Milena,
I do not think that these 2 sequences are caused by the same problem.
Target 1 ( 2 close size bands)
I would run the separation on 1% gel as they are quite large bands and 3% is for resolving very small pcr bands and the extra agarose may interfere with the purification of the dna. Run the gel slow and cold for best results.
The sequencing is good with a strong signal so the amounts of template and sequencing primer are both good. The sequence is single and good up to ggTTCCA when 2 well sequenced sequences appear both sequencing equally well. This is often an insertion/deletion or 2 very similar sequences amplifying together The main sequence looks like REEP1 but without the text sequence of the expected amplimer It is not possible to say what the second sequence is. Can we have the amplimer sequence please? Text or Word. You can also sometimes also get this effect from a target sequence from both inserted and non inserted plasmids but This should not be the case as you have size separated the bands on the gel.
Target2.
To me this looks different. The sequence starts with low intensity signal so the machine is pulling up the background noise and spoiling the quality of the early sequence. The sequence gets better after a while. This late starting of the sequence seems to me to be a result either of too little template dna or too little sequencing primer . Probably the dna amount. Hope this helps get back with any queries.
Paul Rutland thanks for the detailed answer, I appreciate your input. Attached is the text file that contains info for the primer pair I used for target1 (which is indeed Reep1). Meanwhile, I have the results from my latest seq attempt which really improved my signal for target2 but did nothing to the target1 seq... Is it possible that there is an additional product for the primers with the same size but different sequence which does not show up in blast...?
hi Milena Korneck
I had a bad feeling that this was the sequence you were going to send. I had hoped to subtract the proper sequence from the 2 sequences region leaving a single sequence that could be Blasted giving a clue to the problem but sadly the bases in the proper amplimer just do not exist within the sequence that you have produced. It looks to me that you have amplified 2 very similar sequences and both of them are the wrong thing. Cloning and sequencing would clarify this but with similar sequences existing before you design new primers it could be worth trying a touchdown pcr to try to clean up the pcr. The early cycles start well above the annealing temperature of the primers so that when the pcr starts working the template is enriched for a single strand of the highest specificity.
A further thought: as your sequence after ctggtcca then AGCAAAAGACCGAAGTTACGATGCCCTTGTG is the real sequence but we get a mixed sequence of
sequence tcca AG GC CG CG AC TG GC GT G CA T G TG GC AT GT GC
secondary sequence A G C G G C T G T G A T G G C T G
Where 2 bases are listed as sequences. If we subtract the expected sequence from the real sequence then we get AGCG etc as the line above.
Interestingly if we look for that remainder sequence within your amplimer it is very nearly found in this line CAC AGC GGCTGTGATGGCTGCTTCCAAGgtactgtgctggaagtaccacaactacatgag meaning that a chunk of your sequence has been deleted.
This can happen in pcr where a high GC motif finds a similar motif elsewhere in the sequence and a loop forms.
The polymerase reads across the loop and a shorter amplimer is produced. This effect can also be found in the sequencing reaction of a full length amplimer giving a shorter sequence than expected.
I suggest you check the size of your amplimer very carefully and if this is not possible try the pcr with either 1Molar betaine or DMSO to break up the secondary structure. You can use dmso at any concentration up to about 8%.
Similarly if you are doing your own sequencing reaction then you can sequence in presence of dmso or betaine and using a higher annealing temperature than the 50c recommended by ABI Up to 60c Ta should be fine for sequencing high GC regions. If your sequencing is done commercially or in house then warn then that you want it done as highGC sequence. Get back to me about any aspects that are not clear in the above
Last thought is that I slightly prefer the sequencing artefact route to obtaining this result than the poor pcr and an accurate sizing of the product should indicate this.
Paul Rutland thank you! What you wrote about the secondary structures seems to make the most sense to me, I don't have much experience with sequencing yet and didn't think of that. I will try out your suggestions with the DMSO!
Paul Rutland DMSO didn't really do the trick, unfortunately - it did not change anything. Then, I designed some primers with the wild type sequence lying in the messy part of the product. If they bind, I assume the messy part is really a wild type sequence in disguise. A first PCR showed me a product, so I am now exploring, why the sequence got so bad in the first place.
- Badges
- Science method