Ambion’s pSilencer™ 2.0-U6 siRNA expression vector is a plasmid based system that uses the human U6 promoter to drive shRNAs. Although this vector contains the human U6 promoter, we currently use this vector to knock-down genes in mouse derived cell lines and so far the results have been pretty good. Ambion also sells a vector that contains the mouse U6 promoter (pSilencer™ 1.0), however, we started with pSilencer™ 2.0 and since the results were pretty good, we decided there was no reason to switch. I am not sure if we would get even better knock-down with the mouse U6 promoter.
When you purchase the vector from Ambion, it comes with enough pre-cut vector to clone 20 shRNAs. The kit also includes a vector containing a positive control shRNA against GAPDH, a vector containing a negative control shRNA that is not specific to any known genes in the human, mouse or rat genomes, and an annealing buffer designed to facilitate annealing of DNA oligos for ligation. Cloning DNA oligos (ordered separately based on the desired hairpin) into the vector is rather simple: The oligos are annealed using the provided annealing buffer, diluted (extremely important for transformation efficiency), and then ligated into the pre-cut vector. Once the ligation is complete, the mixture is used to transform E. coli to generate single colonies.
Whether or not you get the desired level of knock-down using this vector will probably depend more on the shRNA target sequence than the vector itself (we have been using Genscript and Dharmacon for shRNA target sequences). With a great shRNA target sequence, we can get 90% knock-down using this vector, however, most target sequences will not work this well. I have never compared the same target sequence using two different shRNA vectors, so I cannot say whether or not this vector is better than other commercially available systems. This vector does have advantages which should be noted. The vector is fairly well designed and is relatively small (a little over 3 kb), so it is easy to work with. In addition, the vector uses common restriction enzymes (BamHI and HindIII) for cloning, so you don’t have to order exotic enzymes just for this system. Finally, a unique EcoRI site is located just 5’ of the U6 promoter, so the whole U6-shRNA unit can easily be moved into another vector if needed.
Unfortunately, this vector also has its disadvantages. While some of these can be fixed with a little engineering, others cannot. First, as supplied, it was difficult to clone the annealed DNA oligos into the pre-cut vector. The linearized DNA was not completely cut, so background colonies were a huge problem. I have even had background problems with vector I have cut myself using the correct restriction enzymes. In addition, there is no way to distinguish the shRNA you are trying to clone into the vector from the one already present without sequencing (inserts are the same size so restriction digests do not work), which can be expensive. Another disadvantage of this vector is that you get low yields of plasmid from maxipreps. For this vector, a typical maxiprep yields 100 – 200 g of plasmid DNA in comparison to greater than 500 g for normal vectors. I am not sure if this is a low copy plasmid or if the bacteria have troubles replicating the DNA. Finally, the plasmid does not sequence well in the forward direction, so reverse sequencing primers should be used.
Since we are using this system to generate many different shRNAs, we did some simple engineering to the vector to make the cloning of shRNA oligos much easier. We inserted a random DNA fragment into the BamHI and HindIII cloning sites in order to facilitate the screening of desired colonies from background. This addition allows us to confirm that the initial digestion is complete (size shift of the vector minus the shRNA is not big enough to see on a gel) before using the linearized DNA. This also allows you to screen the colonies before sending them for sequencing. We often use the EcoRI site 5’ of the U6 and the HindIII site after the shRNA. In summary, we like this vector since we have established the system in the lab; however, there might be better systems available that do not have these same types of problems.