Hi Nathalia,

What do you call negative control? I have used 2 types of control for CRISPR experiments: 1- a shRNa targeted to an intron of the same gene so the Cas9 can still induce a double stranded break but in a non-coding region of the DNA. 2- transfection of the empty plasmid. 

The plasmid i used is pSpCas9(BB)-2A-GFP and i have never observed altered expression of other proteins (I controlled for ion channels and trafficking protein expression) than the ones I was targeting (in vitro and in vivo).

Another control used by the Zheng lab is an sgRNA against luciferase. Your observations might be the result of your control having an off-target effect. Another explaination could be that your cancer cells were not in the same conditions of confluency between your transfections.

What do you think?

Hi Aubin. Thank you.

My negative control is  a plasmid with 3 gRNA non specific to any DNA sequencing in human cell. In the wester blotting I added the non transfected cells, the negative control and the knocked out cells and because of this is worrisome my result. The off-target effect is possible by not expected. 

I had this problem with negative controls for shRNAs. When you have a non specific sequence you actually have more chances of an off-target effect compared to when you have a specific sequence (i think the control cannot find a specific sequence to bind, it will bind to the next best thing...). did you try cells transfected with the empty plasmid? are you using a selection antibiotic on your cells? 

We buy a commercial CRISPR/Cas 9 kit to a specific gene. The negative control is included in the kit and because of this we dont have an empty vector. About the selection, I co-transfected the negative control with the HDR plasmid and because of this selected with puromycin.  

Thanks

Another reason could be the effect of antibiotic selection. Maybe you selected for a subpopulation through either your transfection step or the selection step. This could explain why you have these differences in the MAPK proteins you are studying. Also, was the confluency of your cells the same when you did your MAPK assay? cell cycle is important as protein expression is going to change drastically between the different phases.

Hi Nathalia,

I think that in your tumor cells the CRISP/Cas system found the off-targets using control gRNAs. Since the genome of the tumor cell is constantly changed accumulating spontaneous mutations, in your tumor cells these mutations led to appearance of the novel sites that can be used as targets of CRISPR/Cas system. As I know from publications there are software that can predict the possible off-target sites for given gRNA. Find out the sequence of all of the 3 control gRNAs and check them for the off-target sites. If you find such sites have a look what genes' expression thay may affect and compare with the genes that shows altered expression of proteins in your experiment.

Good luck!

Thank you very much Dmitry.

Hi Nathalia,

I do agree with the aforementioned off target effect of your controls. In this particular question, it is good to have in mind that in your guides, the positions 5 to 11 (approximately), called seed region, are the ones that can not tolerate mismatches, the others, however can. I would suggest to make a blast with these seed regions of your guides and see what you find.

The following paper has some nice information:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4338555/pdf/nihms657677.pdf

Good luck and all the best

Hi Marcelo, thank you. 

Very nice article, thanks.

Please see, article PMID- 28474669 , Figure 1 as well.

Hi

Aubin Moutal, Can you tell me please which plasmid you used as negative control and what do you mean by the empty plasmid?

Thanks

   Heba

Thnaks Aubin Moutal for your detailed explanations.

I tried to use LentiCRISPRv2GFP as my empty vector for my CRISPR KO. However, I noticed that it also carries a short guide RNA sequence (5'-gagacggacgtctct-3'). I blasted it on NCBI Blastn and found it also targets a couple of genes (e.g. GIT2). I am not sure if it is the best vector control. I think I may want to use an empty vector which I used for shRNA (pLVTHM) as a negative control. Any thought?

Heba Alaa Eldeen I routinely use these plasmids all available from addgene:

- pSpCas9(BB)-2A-GFP (PX458)

- pL-CRISPR.EFS.GFP

- pL-CRISPR.EFS.tRFP

I clone my gRNAs in these and then use the untouched plasmid as control. Before you clone any sequence in them, the gRNA is not really functional as the existing sequence is too small.

Yung Hsing Huang i think we need to have the cas9 onboard for our controls so the pLVTHM does not seems appropriate as a control in these experiments. The gRNA sequence you provided is rather small and i don't expect it to produce a potent gRNA. Also, if you have any mismatch in the results of the blast you did, then it's unlikely that Cas9 will cut there (people are still failing to prove that CRISPR systems have off target effects). If you are really worried about this sequence you can clone a gRNA for GFP or luciferase and use this as a control.

I hope it helps!

Thanks for your valuable input Aubin Moutal !