293 cells and transient transfection to generate a recombinant Ad stock: I find the most critical factor is the state of the cells.
Firstly, there are lots of different 293 cells out there. Some are great for doing
agarose overlays (because they don't come off when you pour the gel) others are great for hih transfection efficiency. In my experience, the ones that come off really easily when trypsinizing tend to give the higher transfection efficiencies.
Then it seems to be important that the cells do not become over confluent (1:10 split every 3 days). If they are left for longer then the efficiency drops.
Why no FBS for the transfections?
With Lipofectamine 2000 (and other companies) the protocol states that there should be no FBS in the two transfection solutions (one with DNA the other wth the lipofectamine). When these two solutions are added, mixed, it is important that complexes with the right stoichiometry form. Presence of serum messes this up.
After this step, the complexes are added to the 293 cells in a petri which has just had fresh medium added. For the Lipofectamine reagent, FBS is supposed to be absent here too. Then after a certain time (is it an hour or more?) it is possible to add for serum containing medium.
I now use the Fermentas ExGen500 reagent, which I found to be as good as Lipofectamine 2000, much cheaper and allows presence of FBS in the medium on the cells from the beginning. In fact they say that efficiencies are better with FBS. NB. During the complex production stage there must be no serum present.
Recently, after a long spell of not working with Ad viruses I had to re-generate a virus from the plasmid (AdEasy system). I tried 3 times to get CPE following transfection with good cells and ExGen500. The final time I had waited about a week for CPE and nothing came. Then I left the plates in the incubator over a long weekend and by the middle of the next week, just as I was about to chuck them away I saw that they had CPE! Sometimes it takes longer than you would expect...
PM