Unspecific PCR - (May/30/2009 )
Dear friends!
I have big problem with my PCR. I'm making some cloning which doesn't work well by unknown reasons. I have very low percent of colonies, containing my vector plus insert. However, it seems to be not the main problem. When I check obtained constructs by sequencing I'm getting frustrated. I'm discovering frame shifts (extra bp was added), insertions (construct is longer than it should be, sequence at one end doesn't correspond to primer), deletions (construct is shorter than it should be). It seems that PCR is absolutely unspecific and random. However, I'm using high-fidelity "Phusion" polymerase!
The protocol is following:
95 for 5 min
95 for 30 sec
56 for 30 sec
72 for 1 min 40 sec (33 cycles)
72 for 10 min
Length of a fragments I'm trying to amplify is ca. 1200 bp.
I'm really waiting for any helpful suggestions!
I wouldn't be too sure that the PCR is the problem.
Low percentage of colonies and the colonies you obtain have the wrong insert obviously. That speaks more for the insert being a problem.
We had a similar situation not so long ago here as well. It turned out to be the insert wasn't well tolerated in the bacteria we used - my colleague had to use some extra bacteria to get the plasmid going.
Have you analysed your insert sequence in terms of GC composition?
Bomber on May 30 2009, 03:22 AM said:
Low percentage of colonies and the colonies you obtain have the wrong insert obviously. That speaks more for the insert being a problem.
We had a similar situation not so long ago here as well. It turned out to be the insert wasn't well tolerated in the bacteria we used - my colleague had to use some extra bacteria to get the plasmid going.
Have you analysed your insert sequence in terms of GC composition?
The thing I'm working on - to make variants of protein of different length (C-terminal truncations). I'm using pET vector with C-terminal His-tag. Insert, which is coding full-length protein is 1254 b.p. GC content is 39% accordingly to "OligoCalc". Cloning is working really purely. recently I have made construct for full length protein (i.e. 1254 bp), purified plasmid, sequenced it and discovered one extra base pare added at the end of insert which obviously causes frame shift and no His-tage as a result. As I have mentioned in another thread, I have discovered strange thing - vector was digested with Nco I/ Hind III, dephosphorylated. At control plate - 2 colonies, at plate with 1200 b.p. insert - 5 colonies, all contain insert (however,not sequenced yet, so I don't know if insert is correct), at plate with 1218 b.p. insert - >50 colonies, 9 tested, all contain just empty vector. I cannot believe. It seems that cells just have "thrown" away the insert out of vector. How it can be and how I can overcome this problem? I'm struggling over two months already.
DH5 alpha cells are used for transformation after ligation, just to let you know.
Before transforming DH5a, do you check the ligation on a gel?
In case of difficult ligations or many empty colonies I 'try' to exclude that the ligation is poor.
What might help is to setup ligations with vector + insert + Ligase and the same reaction without ligase.
Put both on a gel after incubation and check if you detect differences.
If so, you can usually expect colonies to be positive. It is not a guarantee at all, but might help to detect where the problem is.
Bomber on May 30 2009, 04:22 AM said:
In case of difficult ligations or many empty colonies I 'try' to exclude that the ligation is poor.
What might help is to setup ligations with vector + insert + Ligase and the same reaction without ligase.
Put both on a gel after incubation and check if you detect differences.
If so, you can usually expect colonies to be positive. It is not a guarantee at all, but might help to detect where the problem is.
I have tried once to load ligation mix on a gel, then to cut and purify circularized ligation product (which runs at the top of gel). Then I transformed DH5a with what I have purified. than I have tested colonies. One of 5 was positive, but insert was "wrong". I have no idea what is going on.
Ligation mixture gel images are much clearer if you first heat kill the ligase. Doing test ligations with insert only, vector only, and vector + insert can tell a lot about what is happening. Even more informative is to cut the insert only and vector only ligations with one of the two enzymes (if it is a two enzyme ligation) and running that result on a gel as well. You should get double length fragments if the (other) enzyme is working well. The ratio of double to single length fragments tell you how well the ends and ligation for the other enzyme are working. This doesn't work for "quick ligation" buffers containing PEG, but I never use those. Don't heat kill the ligation product that you intend to transform with. In my experience this reduces the number of transformants.
Givi on May 30 2009, 10:12 AM said:
I have big problem with my PCR. I'm making some cloning which doesn't work well by unknown reasons. I have very low percent of colonies, containing my vector plus insert. However, it seems to be not the main problem. When I check obtained constructs by sequencing I'm getting frustrated. I'm discovering frame shifts (extra bp was added), insertions (construct is longer than it should be, sequence at one end doesn't correspond to primer), deletions (construct is shorter than it should be). It seems that PCR is absolutely unspecific and random. However, I'm using high-fidelity "Phusion" polymerase!
The protocol is following:
95 for 5 min
95 for 30 sec
56 for 30 sec
72 for 1 min 40 sec (33 cycles)
72 for 10 min
Length of a fragments I'm trying to amplify is ca. 1200 bp.
I'm really waiting for any helpful suggestions!
Dear Givi,
It seems that you are not running the PCR in recommended conditions (too low denaturation temp., too long extension etc.) recommended by manufacturer. Please see attached link and at least 'Important Notes' listed in the manual. Those recommendations are really important when using Phusion. I'm pretty sure that the PCR start to work if you use recommended protocol.
http://www.finnzymes.com/pdf/phusion_hs_da...0sl_1_5_low.pdf
Have nice day
Phusion is quite stable at the described conditons as well.
BUT what kind of cloning are you doing? Phusion produces blunt ends, so if you are doing TA cloning you need to add an "A" overhang at the end of your PCR product which should be cloned. If you are doing blunt end cloning: this might be tricky. But you can search for previous threads on this topic, they probably can help.