We are looking for a way to insert DNA into a genome, but we would like to do it in a way that the recognition site stay intact to be able to add again DNA at the same location. Do you know if it is possible or if it is already the way CRISPR/Cas9 does it? Apparently multiple systems were already engineered but the few papers we read weren't about our problem.



  • 1
    $\begingroup$ Inserting into the genome of what organism? $\endgroup$
    – Luigi
    Mar 11, 2015 at 18:13
  • $\begingroup$ Why not just make sure your inserted DNA segment contains the same CRISPR/Cas recognition site, so you can just cut it again? $\endgroup$
    – March Ho
    Mar 11, 2015 at 19:00
  • 1
    $\begingroup$ @MarchHo I had the same thought initially, but then Cas would cut your target DNA as well. Furthermore, the target DNA would also have a region of homology which doesn't contain your desired insert and repair could occur which doesn't result in insertion of your desired sequence. $\endgroup$
    – canadianer
    Mar 11, 2015 at 19:27
  • 1
    $\begingroup$ @EmilieCuillery Is there any reason you cannot just use a plasmid? $\endgroup$
    – March Ho
    Mar 11, 2015 at 19:31
  • 3
    $\begingroup$ The nice thing about CRISPR/Cas is that target recognition is done by RNA, which is relatively easy/cheap to obtain. Could you just get another gRNA which recognizes the new sequence after insertion? $\endgroup$
    – canadianer
    Mar 11, 2015 at 19:35

1 Answer 1


Look into site-specific recombination. You can use a site-specific recombinase, specifically an integrase, that can insert a sequence of DNA at a certain attachment site. You can add an identical attachment site into the DNA sequence to insert, allowing for the reintegration of a new attachment site. Note however that you might get SSR onto the insertion plasmid itself, so don't do this if you need a specific amount of integration.

Look at the bottom subfigure of A for an example of integration.

Look at the bottom subfigure of A for an example of integration. Recombinases can also do other things, such as excision and inversion, based on the orientation of the attachment sites.

  • $\begingroup$ Doesn't use CRISPR though. Right? $\endgroup$ Mar 12, 2015 at 3:17
  • $\begingroup$ It seems that this also suffers from the same issues as pointed out in the comments. If your insert has the recombination site, it may recombine with copies of itself. $\endgroup$
    – March Ho
    Mar 12, 2015 at 7:17
  • $\begingroup$ Our projects is basically based onto recombinase. And work...published 2 months ago ... but not by us :( So we are looking for a better efficiency (10^-4 with recombinase) allowing a faster system. As i understand CRISPR/Cas9 doesn't allow insertion of several piece of DNA at the same place butCas9 is a very powerful tool do you think we could engineer it to cut not the homologous sequence but just after of before? (We really have problems understanding where the DNA we want to insert is coming from and the exact mechanism of it. Are you thinking of another way faster than recombinase? $\endgroup$ Mar 12, 2015 at 8:56
  • $\begingroup$ Anyway thank you all for your help but i think we will abandon the idea for now as we learned in our researches that the Cas9 rate and efficiency is much too low to do what we want... i'm sorry ! But thank you again for your help :) $\endgroup$ Mar 13, 2015 at 9:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .