Heterogeneity of protein expression level in transciently transfected cells.

Frank O. Fackelmayer Frank.Fackelmayer at uni-konstanz.de
Sat Aug 7 04:38:01 EST 1999

Hi Andrew,

Andrew Leung wrote:

> Hi all,
>     I am going to overexpress a protein in my favorite cells. The results of
> transcient transfection lead us to suggest that the level of protein
> expression is critical in affecting the responses of transfectants. Because
> some transfectants die and some remain normal.
>     I would like to ask whether there will be a great variabilty in
> expression level among different transfectants in general?

Yes, the protein levels in transiently expressed cells vary by at least one
order of magnitude. There are several reasons for that. Most importantly, not
all cells take up the same amount of DNA. Some don´t take up any DNA, and the
percentage of cells that take up DNA is strongly dependent on the cell type and
method of transfection. If you use cells that are easily transfected (like COS7
cells), you can get up to 70 or even 80% of cells actually transfected. In case
you have cells that are hard to transfect  (like Jurkat cells), or the method
of transfection has not been optimized, it is not unusual to have less than 1%
transfected cells.
Even in the cells that take up DNA, the amount varies considerably, and so does
the protein level, of course. If the effect you want to study is highly
dependent on the amount of protein made, it will be quite different in
individual cells. If you have a quantitative way to determine the effect (like
enzyme activity), there may not be a problem because you always investigate a
population of cells, and the effect will be averaged. You might even get a nice
dose-response curve when titrating different amounts of DNA. For interpretation
you should keep in mind, however, that transfection efficiency is NOT AT ALL
linear with the input amount of DNA.
In your case, I guess it might be best to make stably transfected cell lines
and select for lines that produce different amounts of protein. Having two or
three lines each of stable transformants that produce, say five different
amounts of your protein (as e.g. judged from western blots of total cell
extracts), you should be able to get results much cleaner than in transient
transfection experiments. Of course, this is not a "fast results" approach,
quite different from transient transfections, and you will have to consider
carefully if the results you can expect are worth all the work to get stable
lines. Even if you´re experienced with that, it will take you months to make
and characterize such lines.

Hope this helps,

> If great
> variability exists and our gene affecting cell in a dose-dependent manner,
> do you agree it is difficult to be shown in transfection experiment and it
> requires other experimentation?
> Andrew

More information about the Methods mailing list