A general question about how HDR functions with respect to repairing a DSB (asked in the context of ‘genome editing’).
I guess this is how HDR might usually be harnessed to insert a desired sequence via HDR…
--------=========|=========--------
V
--------========= =========--------
X
=========++++++=========
V
--------=========++++++=========--------
…but how far can the homology arms be from the DSB for the template to still mediate efficient HDR and generate the desired change? Eg, this -
--------=========-----------------|-----------------=========--------
V
--------=========----------------- -----------------=========--------
X
=========++++++=========
V
--------=========++++++=========--------
Ignoring NHEJ does anyone know how long the intervening stretches could be and HDR still work?
In the MosTIC paper by Robert and Bessereau (http://www.ncbi.nlm.nih.gov/pubmed/17159906) they analyzed conversion tracts by SNP analysis and observed a bell-shaped conversion tract with 50% of the conversions copying up to 1 kb from the repair template. Only 25% of conversions are reported to extend beyond 1kb. That argues that your repair arms should be within 1 kb of the DSB. However, Christian Frokjaer-Jensen’s MosDel work indicates that by keeping homology of the repair template near the DSB and moving the other side away from the DSB that one can delete intervening sequences of up to 25 kb. (http://www.ncbi.nlm.nih.gov/pubmed/20418868)
Am I correct from your diagram that what you are trying to do is both delete a region and insert a sequence? If so, it seems to me that the most efficient method would be to use a CRISPR gRNA to target Cas9 to the left or right end of the sequence that you want to delete and use ~ 1.5 kb of homology flanking the break (===== sequence). Have the other homology arm extend 1.5 kb adjacent to where you want the insertion (other side ====== sequence). Then your ++++ insert should be between the arms along with a LoxP flanked selectable unc-119 cassette. Does that make sense?