I am trying to remove the GFP tag from our construct using the NEB Q5 SDM kit. My primers are designed to flank the region to be deleted and when I look at my PCR products on a gel (before the kinase ligase Dpn1 treatment) they look as I would expect, with the PCR products running slightly lower than the parental DNA. However, following KLD, transformation into XL10 gold and miniprep the products run a full 2-3Kb lower than the parental DNA (I am only expecting a 800bp difference). I have screened a lot of colonies and they are all appearing like this. Also my sequencing results show only the first 500-1000bp of my gene of interest (it’s 4kb) after which it goes straight to the region of the vector after the GFP tag. I have sent 3 different preps to sequencing and there is no consistent length of gene amplified and no obvious other complementary region to which my primers would bind. As my PCR products look fine, but miniprep products are smaller than expected I therefore concluded that for some reason rearrangement is occurring following transformation. I therefore tried to transform into Sure 2 competent cells as these are deficient in some of the pathways that cause rearrangement. However I am not getting any colonies with these cells. My construct plus vector are approx. 14Kb, which may be a little large for Sure 2. Does anyone have any suggestions of what I could try next?