siRNA to shRNA problem - (Jan/23/2006 )

Greetings,

I've designed a shRNA construct from a published siRNA sequence for the knockdown of eGFP. However, when I infected into 3T3s (same cell line as paper), I saw no knockdown of stably-expressed eGFP 6 days later. Is it common for siRNA sequences to have different efficiencies than shRNAs? I can't puzzle this one out. Any advice on how to fix my construct would be much appreciated.

shRNAs are supposedly more efficient than siRNAs, so I'm not quite sure what to make from your results. Did you use a program to design your shRNA from the siRNA sequence?

-Hank

Have you sequenced your construct to make sure sequence is right?

Sometimes the loop sequence has an impact on RNAi efficiency.

How is your transfection efficieny? Have you constructed another control shRNA such as GAPDH and does it work?

Thanks for replying. A thousand thanks for all your helpful comments.

I didn't use a program to design the hairpin. I took the 21nt literature sequence, added the "Ambion loop" from their website, and put it in a retroviral vector flanked by restriction sites.

I have a scrambled hairpin with no sequence homolgy to the mouse genome; that doesn't effect eGFP levels. The shGFP contruct was supposed to be my positive control for RNAi, so I don't have another one on hand.

My infection efficiency is pretty low. I select for infected cells with 10ug/mL puromycin for a little over 24 hours (until control cells dead). I would say less than 40% live-- twice all my cells have died. Could that be affecting RNAi response?

Lastly, I am re-sequencing the plasmids to see if it possibly recombined between cloning and growing up DNA.

Unfortunately I don't have very good answers to your questions. I had assumed that the percentage of cells alive after puromycin treatment was correlated to infection efficiency (or at least infection status). I am using the pSM2 retrovirus, a murine stem cell virus from openbiosystems. I don't know if it provides cellular markers for infection. My construct is orthogonal to the mouse genome, so I'm not concerned about multiplicity of infection.

Since I don't know why the shRNA didn't work, I am repeating the infection process with longer puromycin treatment. The resequenced DNA showed no mutations.

Thanks again for all your help, my advisor is out of suggestions!

i think they're for different purposes. siRNA for transient transfections are better than shRNA as you have an active form of dsRNA. an you can get rid of differences of expression which can add to differences of trnafection efficiency.


I think that these differences are minor regarding sequence design.