"full knockout" but faint protein expressed

[Luciana Madeira da Silva]

Hi everyone,

I am trying to make a knockout for my gene of interest using the CRISPR technology. My strategy employs two sgRNAs about 100bp apart in the same exon, which were cloned into Addgene  #48138 and co-transfected in a couple of diploid cancer cells. I know both sgRNAs have good activity (tested them isolated in 293FT cells). I am using PCR analysis to screen for the deletion mutants. As expected, I had clones that were WT fragment, others were heterozygous (both WT and deletion fragment amplified by PCR) and my lucky ones that show only the deletion fragment. However, western-blot for these two clones still show a faint band of protein expression, exactly at the same size as parental (my PCR did not show any hint whatsoever of the parental fragment). Anyone has seen something similar? Suggestions? I found this with clones of one of my cell lines, I still have to do WB for the knockout clones of my second cell line. I appreciate your comments.

Luciana.

[Charlie]

you may have really low levels of clonal contamination of wt cells. you could clonally isolate the cells then screen again with WB and see if the protein is still there. 

[Eric Billy]

Hello

You might have re-created an open reading frame. We use the same strategy for gene KD, but we design the upstream sgRNA to be either before or very close to the ATG. The downstream sgrNA is chosen to be close to the end of the first exon, in order to remove any chance of recreating an open reading frame after cleavage by Cas9 and religation.

We have seen that too with single sgRNA and decided to move to tandem CRISPR (2 sgRNAs) to remove msot of hte first exon at once, checking that no downstream ATG could serve as internal initiation.

Best

[Luciana Madeira da Silva]

Hi Eric,

Thanks for the reply. I am indeed using two tandem sgRNAs (they are 97nt apart) but they are not close to the start codon. The predicted deletion should not re-create the ORF but as I am sequencing some of my clones I realized that the cleavage site is not exactly as what one would predict and it could be that this happened in the particular clone that I have (I still have not sequenced that one). With the tandem sgRNAs that I am using I found that there is an alternative splicing site downstream, which should yield a truncated protein (54KDa vs. 88KDa for WT). I have a sgRNA designed very close to the WT start codon which I have not been able to validate with the Surveyor assay because my Surveyor primers did not work and the genomic region in question has very low-complexity introns (exon is really short there). But it is a good option, although the downstream ATG could still be a problem. In any case, I really appreciate your time and input, thanks.

[Luciana Madeira da Silva]

Hi Charlie,

Yes, I actually have considered that explanation also, although I would imagine the Surveyor PCR should have shown at least a hint of WT product. But re-cloning is something that I am starting to do. I will keep everyone posted in the progress as I go. Thank you.

[KC]

Hi Luciana,

Any update on this?

Thanks

KC