Amplification of Plasmid Library

Dom Spinella dspinella at chugaibio.com
Mon Dec 7 11:17:29 EST 1998


> I'm making a cDNA expression library in a plasmid vector. 
> I plan to divide the library into 125 pools of about 5000 clones each
> during the amplification stage (this is because I'm screening via a 
> bio assay).
> 
> Methods in Enzymology (Vol152 1987 pp407-415) suggests that 
> the library should be grown in soft agar on LB agar plates to
> minimize differential growth between colonys.
> 
> Whats the simplest method of seperating the bugs from the agar
> (MiE suggests something fiddly involving converting the agar into a paste
> by squerting through a 21 gague hypodermic needle and spinning through
> a Sephadex column).
> 
> Are there any better methods lurking out there?

The easiest way to remove the colonies from agar is just to scrape them
off -- flood the the plate with a few ml of LB and use a plastic scraper
to gently (without scoring the agar) resuspend the bacteria (an old
credit card works well for this).  Alternatively, you can just pipet up
and down, squirting the broth on the colonies until they are
resuspended. Then just prepare plasmid DNA as you would from any liquid
culture.  I fear that if you squirt the bugs/agar through a syringe, you
will lyse many of them and reduce your yield of plasmid DNA.

Now, having said all that, I'm not sure I agree that differential growth
differences are minimized by growing colonies on agar plates rather than
liquid culture. A lot of people assume this but few have tried to verify
it. I tested this hypothesis a few years ago (albeit with M13 phage
clones and not plasmids) and the results clearly show that growth rate
differences remain the same whether the phage were grown in agar (as
plaques) or in liquid culture.  The results are published (McConnell,
Uvegas, and Spinella, "Comparison of plate vs.liquid amplification of
M13 phage display libraries", BioTechniques 18:805-806, 1995).  I
suspect the same holds true for plasmid growth. Anybody else out there
have any data rather than anecdote or "conventional wisdom"?  Cheers.
Dom Spinella



More information about the Methods mailing list