Best cell line for expression cloning but not COS-7 c

Ian A. York iayork at panix.com
Tue Jan 6 14:51:15 EST 1998


In article <34B27569.5FD3 at chugaibio.com>,
Dom Spinella <dspinella at chugaibio.com, @chugaibio.com> wrote:
>
>if anyone has tried expression cloning from a system in which there is
>chromosomal integration.  There might even be an advantage to it -- you

I see two potential problems with this: first, getting enough cells
transfected, and second, getting the plasmid back. 

I gather that CHO cells are total sluts when it comes to transfection, and
so you might be able to get enough integrated plasmids to do the job.
Remember that you need something like a million cells to get adequate
representation of your library.  If the cells integrate at 10% then you
only need to transfect ten million, which is doable.  But most of the
stable transfections I've done (never with CHO, so I may be orders of
magnitude out) give you stable integrants at the level of 1/10,000 or so;
in that case you'd need to start with, um, a whole lot of cells, and it
might not be practical.  I think you'd be pretty reuctant to start with
more than 100 million cells, and you'd prefer a lot less.  

The qualifier here is the number of plasmids that integrate.  Each stably
integrated cell *might* take up more than one plasmid, probably up to ten
or so, which eases the problem slightly.

For getting the plasmid back, you pretty much need to use PCR, as Dom
suggested.

Having said that, I heard about a potentially useful procedure.  I have no
reference for this, and no experience with it, so I'm not making any
promises.  The trick is to take your library plasmids and linearize them
*with a rare cutter*.  (SfiI is potentially useful here; it's in the pCDM8
series of plasmids.)  Then ligate the linearized library.  Transform your
cells with this whole glop of DNA, which contains long strings of DNA
concatamerized.  Any cell that takes up the DNA is therefore transformed
with many, many genes.  Select as usual.  To pull the gene out, pull out
the genomic DNA and then cut it with SfiI, or whatever you used.  Then
ligate again, under conditions that promote self-ligation.  What you get
is your original plasmids back, and they'll replicate in bacteria, so you
plonk the whole pool back into bugs and take the results.  You have to do
several rounds of enrichment, of course.  

I thought the idea sounded really nifty but a lot of work.  

Ian

-- 
    Ian York   (iayork at panix.com)  <http://www.panix.com/~iayork/>
    "-but as he was a York, I am rather inclined to suppose him a
     very respectable Man." -Jane Austen, The History of England



More information about the Methods mailing list