Help finding siRNA targets in MEF and HEK293 cells needed - (Aug/05/2011 )

I'm facing a quite simple task which I have no idea how to tackle, and I thought I'd ask for a few pointers.
I need to find target genes for siRNA in HEK293 and MEF cells. I intend to use them in a knockdown assay where potential inhibitors of RNAi will be present along with the siRNA, and the level of knockdown will help us determine if the given protein is indeed an inhibitor. (That's a first step.)
I'm stuck at the target selection: should I choose a very abundantly expressed gene (like GAPDH), or something that's only moderately expressed, so that the changes in abundance are more apparent? (If there's an inhibitory effect.)
And the main question: how do I know the expression levels of different genes in different cell lines? (Yes, I can start doing Westerns or qPCRs on different genes, but I'm hoping there are bioninformatic tools, databases for this. I just don't know where to look.)
Thank you for your time.

(A bonus question: I used siGlow and all my cells are glowing red in a transfection. Does that mean I have 100% transfection rate? The level of GAPDH does not drop that drastically -viewed on limited cycle RT-PCR) that you would expect if all the cells have the siRNA working.)

Andras on Fri Aug 5 15:03:11 2011 said:

I am not quite sure what the purpose of your study is. GAPDH probably is not a good candidate because it's expression is really high. You can choose some essential genes such as cell cycle related genes whose knockdown can give you expected phenotype.

Yes, there are databases which provide you with expression data of any genes in different tissues and cells, such as Gene expression omnibus at http://www.ncbi.nlm.nih.gov/geo/, protein atlas http://www.proteinatlas.org/, oncomine at http://www.oncomine.org

If you see red in every cell, you can say that the transfection efficiency is 100%. But if GAPDH is your siRNA target gene, you won't be able to get decent knockdown.

GAPDH is tough, as pcrman pointed out. Don't try actin either, we can't even knock that down with a Morpholino, there is just too much transcript being made. Transcription factors are usually good targets because they are expressed at reasonably low levels and they tend to have high protein turnover rates.

Thank you very much for your help. I suspected that I'm barking up the wrong tree with GAPDH... So now it's off to get something different ordered and tried. I really appreciate your input; I might come back asking again (Probably will.)

I’ve been thinking about the assay itself. The whole point of the assay is to see if a protein of interest inhibits the process of RNAi or not. For that, we thought, we simply knock down a gene (any gene –GAPDH in my case) with siRNA in cells expressing the protein, and see if the levels of GAPDH mRNA change. If the protein indeed interferes with RNAi, then the levels of mRNA should not change much as opposed to the wild-type knockdowns. I’m having doubts if this is a valid approach now...
Most of the papers use assays where they transfect the gene to be knocked down as well –GFP or some other constructs. I can see why it makes sense, but on the other hand, it also complicates things as you have an additional transfection which needs to be properly controlled. And I’m having problems with the stable expressing cell lines that have the plasmid of the protein of my interest. It seems like no matter of the G418 treatment, flow cytometry, the levels of expression in individual cells vary a lot. (It’s mCherry tagged, so I can actually see the protein expression.) Not to mention I cannot get 100% of the cells expressing... And additional construct would make life even more difficult.
So to sum it up: is my approach valid, or should I turn to a different assay?
I’ve been thinking about the assay itself. The whole point of the assay is to see if a protein of interest inhibits the process of RNAi or not. For that, we thought, we simply knock down a gene (any gene –GAPDH in my case) with siRNA in cells expressing the protein, and see if the levels of GAPDH mRNA change. If the protein indeed interferes with RNAi, then the levels of mRNA should not change much as opposed to the wild-type knockdowns. I’m having doubts if this is a valid approach now...
Most of the papers use assays where they transfect the gene to be knocked down as well –GFP or some other constructs. I can see why it makes sense, but on the other hand, it also complicates things as you have an additional transfection which needs to be properly controlled. And I’m having problems with the stable expressing cell lines that have the plasmid of the protein of my interest. It seems like no matter of the G418 treatment, flow cytometry, the levels of expression in individual cells vary a lot. (It’s mCherry tagged, so I can actually see the protein expression.) Not to mention I cannot get 100% of the cells expressing... And additional construct would make life even more difficult.
So to sum it up: is my approach valid, or should I turn to a different assay?