It should not matter though I personally have not tried a ssoligo antisense to the gRNA.  Be sure to mutate a base in the PAM sequence of ssoligo to minimize cutting pre or post HR.  

It does not matter since you iniciate a double strand break, hence both ends are open for repair

Cheers

hi guys, thanks for the answers! 

Joseph, I can't mutate the ssODN oligo, because my repair template is in the CDS and protospacer is UGG (Trp) which cannot be silently mutated. So I want to use repair template that sgRNA (sense or antisense) will not recognize. I am just not sure whether to use repair oligo in sense or antisense orientation with sense sgRNA.

Thorben, to my knowledge it matters when you have an intact gRNA recognition sequence in your template, as this will be also destroyed by Cas9

I am having the same reflexion right now as I start to design my first CRISPR experiment. For sure, if your ssODN is the same strand as your RNA guide there won't be any possible annealing, thus no destruction of your repair ssODN. So I would go for it. 

hi Marie, yes indeed I came to the same conclusion as you. good luck with your CRISPR!

Hi Jakub,

There is a paper on optimisation of the design (Optimization of scarless human stem cell genome editing). The conclusion is complement is better below 90bp, for longer ssODN, it does not matter. "...We similarly tested the ssODNs DNA donor design for Cas9-gRNA-mediated targeting. First, we constructed Cas9-gRNA (C2) targeting the AAVS1 locus and designed ssODN donors of variable orientations (Oc: complementary to the gRNA and On: non-complementary to the gRNA) and lengths (30, 50, 70, 90 and 110 nt). We found Oc achieved better efficiency than On, with a 70 mer Oc achieving an optimal HDR rate of 1.5%..."

I haven't tried that yet, but that is how I am designing my repair template. The destruction is still possible after the editing has taken place, therefore you should modify PAM.

Good luck!

http://nar.oxfordjournals.org/content/41/19/9049.full

Dear Iwona,

thanks for the answer and paper suggestion. PAM site I can't modify, because my sequence for multiple sgRNAs is T[GGG). And unfortunately first G I can't change, because TGG codes for Trp that is in my ORF and has no alternative codon coding. 

So I am hoping that by the time HDR happens, there is no Cas9 left to make damage...

I agree with Marie that your sequence should be sense to the gRNA but since HDR already occurs so infrequently (I don't know what cell line) the chances of HDR occuring and not still being cleaved by Cas9 seem pretty slim.

Are there any other gRNAs you can use? I think it would be worth doing even if the quality score is lower and the DSB is a little further. Update us if your expt works!

Hi guys,

so I am back with a update. At the end the strategy worked out, but inefficiently though. I could get my HDR-repaired clones, but I have also detected many clones that had both HDR-intoroduced mutation and also NHEJ mutations. I guess first it was HDR-repaired and was subsequently cut again, since gRNAs could still recoginze the locus and Cas9 was still around...i have sequenced 328 single-cell colonies and got 4-independent mutant clones, all homozygous. Wendy, you are right, it is better to choose different guides. 

Hi Jakub,

good job!!

Just a little tip: Not sure how you determine homozygous or not, but make sure to run a PCR over a big enough region. I often found that the other allele had big deletions (about 1-2 kb) hence leaving the PCRed modified allele looking like a homozygous in sanger sequencing!

Don't want to discourage you, but maybe run a quick quantitative genomic PCR on your gene. I usually run RT PCR for genomic DNA and determine copy number compared to unmodified cells.

Good luck and fingers crossed!

Hi Beeke,

many thanks for the input! My PCR used for sequencing was 0.5kb, so I cant exclude the option you are mentioning. I have primers to amplify the whole gene (~5.5kb), so I will run a PCR on my modified/parental cells to check for the large deletions. 

Anyway, I have not come across this phenomenon in the literature, did you?

No I haven't either. I came across it at the CSHL Crispr/cas9 meeting last year. There was a poster on it and I decided to check in my clones and it seems to happen more often than you think.

Hi Jakub, 

I, too, am using CRISPR/Cas9 for mutation correction through the use of an ssODN. I'm glad to hear that you were able to generate results. I'm running into problems with establishing monoclonal cell lines upon CRISPR/Cas9 editing due to the fact that a selection marker can't be included with a ssODN construct. Have you been able to circumvent this issue and establish monoclonal cell lines or are your clones in a heterogeneous culture?

All the best, 

Chloe 

Hi Chloe,

yes, I have established clonal cell lines without the selection marker. The biggest challenge for me was to optimize the conditions for single-cell growing. 

1. transfected my cells with Amaxa nucleoporation (included also NHEJ inhibitor Scr-7, but  it didnt help at all, maybe because of low concentration)

2. let them recover and grow as a bulk

3. sort single cells by FACS into 96-wells. this was the bottleneck for me, but at the end it worked out with using 50% conditioned media from the same cell line and pre-coating the wells with type I collagen. I also thought about using some irradiated stromal cells as growth support. 

4. wait for single clones to grow, replicate the growing wells.

5. isiolate DNA, PCR and sequence

Thank you for your response. We have had no success with culturing single cells, so I'm glad to hear that this method works. 

i think it is very dependent on the cell line and for sure easier with tumor cell lines.

Hi Fatima,

I guess Beeke by RT-PCR meant real time PCR on gDNA and she runs it for copy number to check for large deletions, not for confirming the clone genotype.

to your questions:

- I used 100nM Scr7 according to the Maruyama Nat Biotech 2015 paper,but others used 1uM Chu VT nature Biotech 2015. I guess you can go up to 10um as long as you see no toxicity, but I see you use 20uM, should be sufficient :-) I guess low concentration of Src7 in my case led to no improvement on HDR

- I have tried 2 different cell lines Hep3B and HepG2 and at the end the same protocol worked only for Hep3B. compared to hepg2, these are tumorigenic and p53 null, so they have increased survival compared to HepG2. 

There is a paper from Richardson et al., 2016 'Enhancing homology-directed genome editing by catalytically active and inactive CRISPR-Cas9 using asymetric donor DNA' suggesting that asymetric ssODNs complementary to the sgRNA can yield up to 60% HDR-mediated repair.

When performing homology-directed repair (HDR) mediated gene repair with CRISPR-Cas9, the choice between a sense or antisense single-stranded oligodeoxynucleotide (ssODN) template is important. The ssODN acts as a repair template that the cell's machinery uses to repair the DNA break induced by the CRISPR-Cas9 system.

Here are the key considerations:

In summary, while the general guideline is to use an ssODN that is complementary to the non-target strand (often antisense to the sgRNA), the optimal choice can depend on various factors, including the specific application, target site, and cell type. Therefore, it's advisable to perform initial experiments to determine the most effective orientation for your particular scenario.

l Take a look at this protocol list; it could assist in understanding and solving the problem.