Is it a must to mutate PAM sequence?

[Jerry Xu]

Is it a must to mutate PAM sequence?

Will Cas9 continue cutting the HR-repaired new strand carrying our desired mutation?

My problem is I'm actually studying some noncoding sequences that we know very little about. I can't easily introduce any other mutation just in case what I introduced may be functional.

Hope to listen to your experience.

Many thanks

[Anoeska]

I've done some things without changing the PAM, but efficiency gets really low (~1% I had in cells) and you have to check if your genomic insertions are without indels (because this can also be cut)

If you can 'scale up' things enough and check really carefully, it's possible.

Op donderdag 1 oktober 2015 03:54:55 UTC+2 schreef Jerry Xu:

[Joe Miano]

Without question a must.  We got burned with a mouse modeling a human SNP where we did not mutate PAM (because then we are not accurately modeling the SNP!).  We got the intended SNP but also some upstream deletions, meaning the mice are useless.  I agree, with enough screening one could get buy w/o the PAM mutation but it becomes cost prohibitive when working with mice.  Incidentally, we just genotyped a mouse with a PAM mutation (NGG > NGC) but still, we saw the intended mutation PLUS a deletion.  SO, apparently, even when the PAM is mutated, one could get intended and unintended edits.  This likely occurred with an initial NHEJ of the region followed quickly by HDR repair with the ssoligo carrying the PAM mutation + desired SNP (the deletion was just outside the homology arm of the HDR template!).  All things are possible !

[Jerry Xu]

Hi I'm quite curious, why would you acquire some deletion in addition to your desired SNP? Where's the deletion from?

Problem is: If mutate PAM, then you are introducing a new mutation, right? This will also be useless to me.

Esp I'm working to see the effects of noncoding sequences for which we know very little about. But if you guys are working on coding sequences trying to introduce nonsynonymous mutations and introduced PAM mutation is only synonymous, well, maybe I'll buy it (but remember even exon synonymous mutation play a huge role in splicing)

[Dale Cowley]

We have had success in mouse embryos introducing SNPs without mutating the PAM as long as we use a guide RNA that overlaps the SNP such that the introduced SNP will be mismatched relative to the guide RNA. The presence of the mismatch reduces cleavage sufficiently that you can get the SNP in. However, if you don't have a guide RNA that overlaps your SNP it is going to be difficult to get it in without also getting other indels.

[Fatwa Adikusuma]

I agree with Dale, I have also generated mice with point mutation without mutating the PAM, as long as the gRNA can't recognise the changed sequences. I also usually create silent mutation to create restriction site for screening and mutation which will additionally create more mismatch in the gRNA target.

cheers,
Fatwa

[Joe Miano]

We find 1 base pair changes are not enough to prevent subsequent NHEJ even when overlapping seed seq.  Again, enough screening will often reveal the single SNP mutation, but it can be cost prohibitive unless you have your own injection station and injectionist.   Agree about the silent mutation in PAM.  

On Wednesday, September 30, 2015 at 9:54:55 PM UTC-4, Jerry Xu wrote: