Crispr cas9 protocol.for generating cell line with three gene knockouts?

What organism are you working with? I work in yeast (haploid) and have done multiple sequential modifications in a single strain. I use the DiCarlo et al two plasmid system, one containing a LEU2 marker and galactose-inducible Cas9 and the other with a URA3 marker and the gRNA expression cassette. These are on AddGene (#43803 and 43804.) Steps are as follows:

Transform the Cas9 plasmid into your target strain

Build the gRNA expression cassette using 5-piece Gibson assembly, transform in E. coli and sequence-verify

Assemble donor DNA cassette (PCR). Verify on a gel or by sequencing if necessary.

Induce Cas9 expression in your cells with Galactose for 2hrs. Transform the gRNA plasmid (1ug) and donor DNA cassette (6ug). 12-hour outgrowth in non-selective galactose medium. A good positive control is using the original plasmid (43803) which contains a gRNA for canavanine. Plate on canavanine-containing plates and look for survivors.

Plate on CSM-Leu-Ura (this selects for cells that have taken up the gRNA plasmid.) Growth may take 4 days.

Colony PCR of 3-10 colonies. For large insertions or deletions, the efficiency is much lower than for few-nucleotide changes. For few-nucleotide changes, verify by sequencing. For large insertions or deletions, colony PCR with a change in band size or presence/absence of a band.

If you don't get any positive transformants, look at the sequencing data. If you see weird indels near the cut site, they are likely due to Cas9 cutting followed by NHEJ repair, which indicates that the problem is with your donor DNA. If there are no weird indels in any of your sequenced colonies, there's likely a problem with your gRNA (it may be low efficiency.) Because the Cas9 itself is only expressed under galactose induction, it's unlikely that there's selective pressure for the Cas9 to be mutated, but if you're unsure, start with a fresh, sequence-verified Cas9 plasmid.

Once you find a successful transformant which has the correct genomic change, plate the cells on CSM-Leu+5-FOA to remove the gRNA plasmid but retain the Cas9 plasmid.

If you have a protein-based or fluorescence-based assay for the change you're trying to make, verify the phenotype at this point after a single passage of your colony.

Repeat steps 2-7 with each additional mutation you want to add. If everything works perfectly, this cycle takes about 1.5 weeks per mutation added.

Alternately, there are apparently systems capable of expressing multiple gRNA cassettes at once. I haven't personally worked with these systems, so I can't vouch for their efficiency.

Sarah Stainbrook

By the way, OU was my alma mater. ;)

Evgeni Frenkel

you can use gRNA vectors with different drug resistance cassettes: puromycin, blasticidin and hygromycin are a good set. Conservative approach is to infect and apply drug selection for each vector separately but may save time to combine or rearrange the steps somehow. Note hyg selection is slow compared to puro and blast, may take up to a week to kill all sensitive cells. I would isolate clones for confirmation of knock-out at the end of all three selections rather than at intermediate stages.

Here's a version of pLentiCRISPR with blasticidin instead of puro

http://www.addgene.org/83480/

Better if only the first vector expresses Cas9 and the other two express only guides. Here's a lenti vector for expressing guides (called sgOpti). It has puro, which you can replace with blast and hyg to get all the vectors you need.

http://www.addgene.org/85681/

An alternative approach is to clone in different fluorescent proteins instead of drug resistance cassettes and isolate cells positive for red, green and blue by FACS.

Pratik Ramjivan Shriwas

Hi..I would like to discuss my method here. Please suggest if I am wrong somewhere. I am following Dr. Feng Zhang protocol. https://www.addgene.org/crispr/zhang/faq/

Pratik Ramjivan Shriwas

@Sarah: I am working with bacteria and I want to generate cancer cell lines expressing only one type of specific protein.

Evgeni Frenkel

The approach suggested by the Zhang Lab faq is obviously fine if you're working with a cell line having high transfection efficiency.

Pratik Ramjivan Shriwas

@Evgeni: Thanks. I am working with A549 nonsmall cell lung cancer cells. I want to knockout three proteins so that is the main concern..........

Evgeni Frenkel

There are a variety of transfection reagents out there (e.g. PEI, XtremeGene, TransIT), some of which may work very well with A549. You can also suspend the cells and electroporate them (e.g. search nucleofection), which always works but is more of a hassle. People in labs that routinely use A549 will surely know the best ways to get DNA into them. Otherwise I think you've gotten here all the general information you need to figure out how to make your triple knock-out. Good luck!

Immunohistochemistry - How to do it?

Is 65% sequence identity good enough for basic modelling using Modeller?

Can docking results from autodock using vina be published?

Software for homology modelling?

How to removes excess blocking?

Crispr ko cell line?

How can I identify metabolites from LC-MS/MS data?

What is wrong with my transfer during western blot?

Is my blot publishable?

Tissue origin of HOP92 cells?

MD MOE RMSD

MD MOE RMSD

fatal error

fatal error

fatal error

fatal error

Can we classify HV and LV lines based on line length?

What kind of fluid could I use for a pitfall trap that also does not invalidate molecular testing?

No title

How to repeat an operation on an atom selection for multiple times in PyMol?