I performed CRISPR/Cas to knock out an adaptor protein from two established cell lines (mouse and human). I performed multiple rounds of sequencing of the segment of target exon where the guide sequence was supposed to bind and confirmed multiple indel mutations (for the different alleles).
I confirmed via western blotting 'almost' negligible protein expression. There were very faint bands, but I and my supervisor blamed it on the background from the shitty antibody. We always thought that CRISPR/Cas is supposed to give a Yes or No knock-out result, nothing in between. Meanwhile, recently, in Mass Spec results we found out that the cells still express this protein. This has left us baffled. While we know the gene for this protein expresses various different isoforms due to the presence of multiple internal promoters and splicing variants, I can't figure out if this leaky expression is due to a different(shorter) isoform that I couldnt target or has something to do with the background I see on my western blots.
You did not say whether your KO lines were clonal. If not, it is likley that you have a heterogenous mixture with some cells escaping the KO.
CRISPR KO of proteins can be very challenging. Since it sounds like you are targeting an exon. Multiple things could be happening. First you are targeting an exon so this will likely not affect transcription of the protein meaning that at the very least the RNA will still be expressed.
Since the RNA is expressed the indel mutations you created could keep the protein in frame meaning the roughly full length protein will be translated and detected by western/maspec.
This doesn't mean the protein is functional but it is detectable.
What we did in butterflies is make guide RNAs to the coding sequence that coded for the domain of the protein that was related to the function we wanted to KO.
Since we knew there was a good chance we'd get small indels we wanted to have the best chance of disrupting the functional domain of the protein for our desired assay.
One thing you could try is making guides to the promoter to disrupt the expression or use multiple guides to completely remove exons.
Or is there a way to test the protein's function in your cell lines?
You did not say whether your KO lines were clonal. If not, it is likley that you have a heterogenous mixture with some cells escaping the KO.
First you have to do the clonal selection after CRISPR/Cas9 transfection. after clonal selection genotype them for knocked out region. if you have done the same, identify the cell genomic condition for any polyploidy. There are various cell lines which has polyploidy, due to CRISPR/CAS9 system does not work.
In CRISPR/Cas9 KO system, you must need to pick monoclonal cell line. Then, you can easily confirmed whether KO occurs by WB, IFA and/or mRNA quantification....
I have a similar issue here as well. However, I am using the base editing system on bacterial cells. I'm not getting any base editing but i'm not sure why. I'm quite sure the plasmid has been successfully transformed into the bacterial cell but I keep getting negatives. I've even checked my design and and my experimental procedures but I'm not sure what's wrong. Does anyone have any suggestions or is there anyone I can consult regarding this matter?
Hi Rachel,
Is there a way you can verify that the Cas is expressing in your bacteria? Maybe by Western blot? Is your guide RNA on the same plasmid? Do you know if that's expressing?
Finally Cas proteins are essentially bacterial immune systems. Has anyone edited your strain before? Do you know if you are using the correct Cas protein for the strain of bacteria?
Good luck!
Christopher Day Hi, Christopher! Thank you for your inputs. Hmm... I haven’t tried running Western blot. But that might be a bit of a problem because I will need antibodies targeting the Cas protein, if I’m not wrong.
Also, I’ve done the same experiment on this strain before and it worked back then. For this current experiment and the previous, i used different mutant versions of the same strain.
Is there any other way I could check if the Cas is being expressed apart from Western Blot?
I’m terribly sorry if it’s confusing for you.
True. There should be some well validated Cas9 (I'm assuming you are trying that protein) antibodies. Some Cas proteins are also tagged so an antibody to that epitope should work as well. We are working with a Flag tagged Cas9 so an antibody to Flag would pick up the Cas9.
But it sounds like you have the system working in at least one strain. Maybe the guide RNA isn't very affective. You could try multiple guides to the same area? Deletions can be hard to screen for if they end up being very large. knock-ins can really depend on the homology arms.
Maybe it worked and the screening method isn't detecting it?
Christopher Day ah I’m actually using a nickase Cas9 fused to a base editor. So my detection method would be to extract the genome, run PCR around my supposedly edited region and send it for sequencing. Plus, it’s quite difficult to find multiple guides because I’m targeting a base and a region that’s very specific. I’m lucky enough to find even one guide RNA to target the gene of interest haha.
Anyway, yeah it seems like there’re loads of confounding factors and many aspects to consider. Maybe my design is flawed considering how I designed the sequence “manually”. I did verify the design on this software called Benchling though and the design seems fine.
Rachel,
Sorry, I hadn't realized that the base edit system is for doing very small/point mutations. I guess you really are limited in guide design. Benchling is good. I just used the Broad Institute GPP and it was very helpful.
When we were doing manual guide design we attempted to find GGNGG sequences because the GG after the PAM seems to help with specificity. We also looked for 50-70% GC content... But as you said you're very limited in available guides.
If you really can't get it to work with the base edit have you thought about knocking in? Cut twice and add in a sequence with homology arms and the desired mutation you want to generate? That might end up being a lot more work...
Christopher Day Yeah, I guess I am kinda restricted in my options haha. Hmm maybe I'll give the Broad Institute GPP a go just to try it out. Thank you!
Also, this is interesting. Are there any papers supporting these observations or did you and your team just recently found these in your experiments? This could perhaps help me find some sort of explanation as to why my experiment doesn't seem to be working as effectively.
And, yes. At this point, we're even considering the knock-in method or just using the regular CRISPR to knockout genes. I'll see if I can work with this.
Thank you so much!
We were working with some older papers.
GGNGG
Farboud, B., and B. J. Meyer, 2015 Dramatic enhancement of genome editing by CRISPR/Cas9 through improved guide RNA design. Genetics 199: 959–971.
Matthew L. Schwartz and Erik M. Jorgensen. SapTrap, a Toolkit for High-Throughput CRISPR/Cas9 Gene Modification in Caenorhabditis elegans. GENETICS April 1, 2016.
Reviewed in:
El Mouridi et al., Reliable CRISPR/Cas9 Genome Engineering in Caenorhabditis elegans Using a Single Efficient sgRNA and an Easily Recognizable Phenotype. G3 (Bethesda). 2017.
- Badges
- Science method