ssODN design - which strand to use?

[Emma]

Hello,

I am trying to make a point mutation change in my target sequence with a single-stranded donor oligo. Does anyone know which strand is better to use for the ssODN?

My understanding is that it depends on whether your CRISPR/Cas9 cut site is upstream or downstream of your target site because the DSB results in 3' overhangs (due to repair machinery processing) and if you have only a single-stranded homologous donor you will have more efficient editing in one direction than the other. Has anyone heard or read something similar?

Thanks!

Emma.

[Matthew Thompson]

Hi Emma,

I'm curious if you've found an answer to this question. I'm making a single point mutation as well and trying to get all of the finer points of ssODN template design down. I've read a couple things about this specific question, but I'd like to hear what you have done with it since it's been over a year since you asked!

Matt

[Hadi Bayat]

Dear Emma

 

Hello,

At begging, I should recommend you to utilize the double nickase strategy instead of the Cas9 wild type form owing to fact of achieving high accuracy. Do you think is there any difference between sense or antisense strands?! You are going to induce a point mutation at desired locus, so the most important thing is designing the ssODNs pertinently containing silencing mutation at PAM site (NGG to NGT, NGC or NGA) and some silent mutations at target sites for avoiding gRNAs binding after introducing your desired point mutation. Kindly see Inui et al (2014)”Rapid generation of mouse models with defined point mutations by the CRISPR/Cas9 system".  For ssDNA oligo design, the Zhang lab typically uses around 100-150bp total homology. The mutation is introduced in the middle, giving 50-75bp homology arms. The Zhang lab typically use PAGE purified long oligos. And it is best if the cut site is as close to the junction of the homology arm as possible. More preferably should be less than 15-10bp away.

 

Good luck,

[Joe Miano]

All
If the goal is to model a human SNP, I strongly recommend only making the desired mutation. We have excellent luck using the sense ssoligo to which the gRNA will not anneal contains just the mutations of interest. And no silent mutations which could have unknown effects. Doing Otherwise defeats the whole purpose of precision guided editing

Just screen screen screen. You'll get what you want
Joe

[Hadi Bayat]

Dear Joe

Introducing a silent mutation at PAM site is a must!!!

[Joe Miano]

Hi Hadi,

We can discuss more in depth off line but I want readers to understand clearly the following.  It is indeed possible to generate mutations without altering the PAM sequence as the attached paper of ours clearly demonstrates.  Second, the introduction of extra mutations in the HDR template defeats the strength of genome editing for modeling SNPs.  Even if the extra mutation is silent, there are reports (e.g., Science paper in ca 2007) showing how such silent mutations can have unexpected effects on expression , translation, post-translational processing etc due to less optimized codons .  As a rule we do not mutate the PAM and we use sense ssoligos to which the sense gRNA cannot bind.

Joe

[Hadi Bayat]

Dear Miano

Hello,

Thank you for sharing your fascinating paper. I read your paper. Yes, you right. You didn't introduce any silent mutation at the PAM site. Instead you introduced 5 mutations at crRNA's seed (12 nucleotides immediately 5' of the PAM site). These days, there are 2 strategies for avoiding the gRNA binding at desired loci: 1- silent mutation at the PAM site, 2- mutation at the crRNA's seed. You utilized second strategy. So we can conclude that choosing the sense ssOND has NOT any effects on  avoiding the gRNA's binding. In fact, the 5 mutation at the crRNA's seed played key roles. Thank you in advance.

With kind regards,