Protocol Online logo
Top : Forum Archives: : Molecular Biology

colony formation assay - (Jun/11/2005 )

Hello there,

I want to do a colony formation assay in order to demonstrate the oncogenic feature of my gene.

I thought I have to generate stable cell lines overexpressing my gene of interest then use the stable cell line for colony formation assay. I saw a paper published recently where it used transient transfected cells for colony formation assay. What is the norm? Do I have to use stable cell line or is it OK to use transient transfected cells?

Thank you~


Personally, I have only used stables. I would focus on making the stables as
there will undoubtedly be lots of useful experiments you can do with cells that overexpress your
gene apart from the colony formation assay.
While it may be possible to do a colony forming assay with transients, several variables which will need to be taken into consideration are the % of cells which are actually transfected and how long transiently transfected cells actually express the exogenous gene versus the time it takes to grow the cell colonies (on soft agar I presume).
It has been my experience that the shortest route to answer a question is rarely, if ever, the
best one.
Thats my 10 cents.


I'm with Mik, make a stable cell line. Its not that hard and doesn't take that long and gets rid of a lot of variables you don't want to deal with.


Thank you so much for the reply!
I think I will try to make the stable cell lines.
Another question relating to stable cell line: the antibody to my gene/protein did not work on westerns. So I won't be able to use western blots to choose the clone expressing protein at highest level (the antibody works on immunofluorescent but it is difficult to quantify the intensity of signal using IF).
I was thinking would it be all right if I epitope-tagged my gene (e.g. myc or flag) and use this plasmid to make a stable cell line, then I will be able to use commercially available anti-myc or anti-flag antibody on westerns to pick the best clone? So the question is whether I can treat this 'fusion protein' as my protein of interest (I am worried that the result will be criticised that I am testing the oncogenic feature of the 'fusion protein' and not my protein itself). Any ideas/comments?