Establishing stable cell lines...

ladasky at ladasky at
Thu Sep 14 20:07:52 EST 2000

In article <099E12FB7F62D1119CDB00805F1521CA0139D973 at TELLY>,
  EGauthier at ("Gauthier , Eric") wrote:
> I am trying to establish stable mammalian cell lines
> expressing a transfected gene by G418 selection in CHO
> cells.
> I treat the transfected cells with G418 @ 400ug/ml (I
> assume this is this enough ?) but don't see extensive
> cell death or gain much expression after 2 weeks
> selection.
> A few things come to mind:
> 1. In order to ensure that only transfected cells are selected, you
> perform a dose-response experiment  with G-418. Try to cover a range
> G-418 that extends from 50 to 1500 ug/ml (this can be easily done in
> plate and using a spectrophotometric method- like the MTT assay- to
> cell viability). Some cell lines require much more than 400ug/ml G-418
> effective selection: if you don't get this one right, you will never
> your clones...

Absolutely.  I have worked with mammalian cell lines for which the
correct dose of G418 is about 200 ug/ml, and others for which the
correct dosage is about 1500 ug/ml.

I did not titer G418 by MTT assay, although that might work and be less
time consuming than my method.  Instead, I actually counted live cells
on a hemacytometer.  Of course, you need to be working with a suspension
cell line to do it my way.  Are CHO cells adherent?

I placed untransfected cells in wells on a 48-well plate, in medium
supplemented with varying doses of G418.  Periodically I would resuspend
the cells, take an aliquot, and count them.  I did this four or five
times over the course of two weeks, and fit a logarithmic growth (or
decay) curve to the data.  Medium was exchanged as needed.

I did this assay for a full two weeks, because that is the typical
amount of time that one needs to see transfectants growing out.  Three
or four days of data will probably be inaccurate for this purpose.

My G418 doses were 0, 100, 200, 400, 800, 1200, 1800, and 3000 ug/ml.  I
was informed that excessive G418 concentrations were as detrimental to a
successful experiment as an inadequate concentration, therefore I kept
my step sizes small (a factor of 1.5) in the middle range.

For my transfection experiments, I chose the first G418 concentration
for which the growth rate of the untransfected cells was clearly

> 2. Cell death with G-418 is usually complete after 1 week or so.
> Could the problem be that I use Lipofectamine 2000 and
> get so much DNA into the cells that too many of them
> remain resistant to the G418 ? Might a less efficient
> transfection protocol like calcium phosphate work
> better in the long run ?

Unfortunately, I'm not familiar with lipofectamine.  I transfect by

Good luck!

John J. Ladasky Jr., Ph.D.
Postdoctoral Fellow
Department of Biology
Johns Hopkins University
Baltimore, MD

Sent via
Before you buy.

More information about the Methods mailing list