Growing large plasmid (20kb) in e.coli - Low plasmid yield from large construct (Jun/29/2005 )
I have been havinga terrible time growing and isolating ~20kb DNA plasmid in E.Coli using stratagene's XL10 Gold Ultracompetent cells and purifying with qiagen mini kits. When I get any yield, it is never more that 15ng/ul. I eventually need 100ug of this DNA. Any suggestions? Thanks.
Is your plasmid a low copy number? If so, good luck 
Reguardless, you can hope for a better bacteria yield by using Terrific Broth or Super Broth.
I''m not at the lab right now so I don't have the recipie, but I'll try to get them for you tomorrow.. or you can consult Maniatis/ Sambrook.
Check the miniprep instructions. Many of them, especially those with silica binding, will fail for large inserts, or fragment them. You might think about the pSCANS series of vectors, which are copy number inducible. You can clone into single copy plasmids and then induce to high copy number for preparation of DNA.
The vector is something old (from th early 90's - pGEM 11zf) that I was given to work with. I would imagine that it is NOT a low copy number plasmid. Can you recommend a few? Then do I just ligate my plasmid into the backbone of a low copy number plasmid and hope for the best?
Thanks
Do you know where I can get the pSCANS vector(s) and their maps? I did a quick google search but didn't find anything immediately helpful. Thanks.
I think you need to do this the old fashioned way. Try to prep your 20kB vector via CsCl gradient. This is a much better method than Quigen for purifying large vectors like Cosmids and YACs. There is a protocol in Manniatis. I would also suggest as someone already has that you use Terrific Broth instead of LB. Dump the Qiagen for this one and you will get a better yield.
Cheers,
LTR
First hit on google for pscans:
http://www.genome.bnl.gov/Protocols/pscans.shtml
I have the same experience with large plasmids, but then with the Macherey nagel Nucleospin kit. In my case, the problem is the composition of the column that has low binding capacity for the large constructs. I use a modified version of the Supercos 1 vector. When I need the construct in large amounts, I do a midiprep with the promega midi kit, and use the low copy plasmid protocol (100 ml culture per column) and then get a yield of about 1 ug/ul.