Hi! I tried this method on almost 20 cell lines (all of them small cell lung cancer cell lines). I used a pTRIPZ backbone, which is an inducible system where RFP is expressed together with an shRNA or a cDNA. Therefore, I observed RFP positive cells just after the shRNA/cDNA induction. I started puromycin selection and maintenance in 2.0 microgram/mL. If the shRNA worked only partially (as in your cases), or if the amount of the over-expressed protein was not sufficient, I moved to 4.0 or 8.0, also 16.0 puro in one case (if I obtained a resistant sub-population in several weeks). Then, I repeated the WB. I obtained significant improvements of shRNA efficiency or cDNA expression in just one case, when I started from a population of cells where I had a clear heterogeneous RFP expression (some cells high RFP, some very low). After moving to much higher puromycin, just the cells with high RFP induction survived, and I obtained much higher cDNA expression. This was the only case.

In my experience, if you already observe GFP positive cells and you start from an almost homogeneous population (I mean, 2-3 times GFP intensity signal between each cell in the population, I checked with FACS), it's unlikely you can significantly improve shRNA efficiency. Maybe you can improve a little bit, but I do not think you'll obtain an 80-90% knockdown.

I suggest you to move to another shRNA, but you can also try to infect with much higher concentrations of virus (10-100x). This can significantly improve the efficiency. If you are not used to concentrate, and you do not have an ultracentrifuge, use the lentiX concentrator of clone tech. It's just a 45 minutes centrifuge at 1500g.

Dear Fernanda, 

Why do you think your shRNA is more effectivce than 40-60%? Is it a comemrcially available and widely tested construct with known high efficacy in your cells? I personally find this knockdown efficiency sad but normal... 

I personally do not work with shRNAs (I use siRNA duplexes only), so I cannot really tell if increasing puromycin concentration will really help. But I don't really see why would it. 

Why don't you just try? Even siRNA-transfected cells suffer the knockdown effect for days (for at least 7 days, to be precise), so there is no real problem with sub-culturing the cells, I guess. 

One more question - did you only try one shRNA against your target? 

Yes it would increase your knockdown efficiency but at the cost of lower cell growth due to increased selection condition 

you will then have to allow for sometime for the cell to reach confluency

When I do shRNA transfection I select at 5X the ic100 conc. to get maximum expression of my shRNA then I clone those out and confirm the clones.

I have been very successful using this method (85-97% knockdowns) using just single hairpins.  Alternatively use a mix of several hairpins to boost your knockdowns,   

Thank you All!

I have used lentivirus to infect my cells with the shRNA I am interested. I use puromycin for selection. All my cells express my plasmid, since all are GFP positive.  I am not transfecting every time, since I have chose in the past to have a stable KD. I really wish to increase the level of my KD without going after other method. I hope I can "force the system" by increasing puromycin concentration...

Thank you again!

 Hi! I tried this method on almost 20 cell lines (all of them small cell lung cancer cell lines). I used a pTRIPZ backbone, which is an inducible system where RFP is expressed together with an shRNA or a cDNA. Therefore, I observed RFP positive cells just after the shRNA/cDNA induction. I started puromycin selection and maintenance in 2.0 microgram/mL. If the shRNA worked only partially (as in your cases), or if the amount of the over-expressed protein was not sufficient, I moved to 4.0 or 8.0, also 16.0 puro in one case (if I obtained a resistant sub-population in several weeks). Then, I repeated the WB. I obtained significant improvements of shRNA efficiency or cDNA expression in just one case, when I started from a population of cells where I had a clear heterogeneous RFP expression (some cells high RFP, some very low). After moving to much higher puromycin, just the cells with high RFP induction survived, and I obtained much higher cDNA expression. This was the only case.

In my experience, if you already observe GFP positive cells and you start from an almost homogeneous population (I mean, 2-3 times GFP intensity signal between each cell in the population, I checked with FACS), it's unlikely you can significantly improve shRNA efficiency. Maybe you can improve a little bit, but I do not think you'll obtain an 80-90% knockdown.

I suggest you to move to another shRNA, but you can also try to infect with much higher concentrations of virus (10-100x). This can significantly improve the efficiency. If you are not used to concentrate, and you do not have an ultracentrifuge, use the lentiX concentrator of clone tech. It's just a 45 minutes centrifuge at 1500g.

I agree with Francesco!

The promoter of GFP is different from the promoter of shRNA of interest (generally H1 or U6), hence force selection for puromycoin may not exactly result in optimal downregulation as this depends on the promoter of shRNA, integration of lenti vector, sequence of shRNA and the intracellular processing of shRNA. I used to optimize the shRNA and then select the best for the Lenti vector, even then, getting 80% downregulation will be a probability game in lenti expression system. The only option I guess in this case is Tet-On/Tet-Off induction.