I have been having a problem with the cloning of a synthesized construct (6.7kb, provided in pUC57 from Genscript company) into my desired vector (~8kb). I first cut the synthesized construct and desired vector with BamHI/EcoRI. Then run the agarose gel and elute the appropriate fragments. I always get the expected band sizes from the first digestion, so I do not think this is an issue.
After elution of bands from the gel using Promega's "Wizard SV Gel Clean-Up Kit", I check the concentration of DNA by Nanodrop and use a 1:5 or 1:3 ratio of vector:insert, with about 10ng of vector for the ligation reaction in a 20uL total volume. We use Invitrogen's T4 DNA ligase at 14C overnight for the ligations. I also do a control reaction with no insert, and do not get resulting transformants (or minimal, only 1-3 per plate).
Then, heat inactivate the ligase at 65C for 20 minutes - or not... sometimes I don't, and I'm not sure it matters. After chilling the DNA on ice, I use 10uL of the ligation reaction to transform homemade DH5alpha chemically competent cells. I culture these cells on solid LB+spectinomycin and streptomycin. The following day, single colonies are cultured in liquid LB media + spec/strep for 16 hours, and then DNA extraction by Sigma-Aldrich's "GenElute Plasmid Miniprep Kit".
All is well until I extract DNA from the cultured colonies: all I seem to get is supercoiled DNA. See the attached picture as an example of the resulting colony DNA extractions. The last two lanes on that gel are the initial construct and desired vector digested with EcoRI/BamHI. The red box denotes the size of the fragments that were eluted and ligated.
For colony DNA digested, I use 10uL of extracted DNA, 1uL each BamHI and EcoRI, 3uL Cutsmart buffer, and 15uL H2O for a total of 30uL.
Why do I get such "random" banding pattern? Any help is appreciated!