Robust Factor IX Expression Following ZFN-Mediated Genome Editing in An Adult Mouse Model of Hemophilia B

Abstract 668

As a therapeutic strategy, site-specific modification of the genome has the potential to avoid some of the disadvantages of traditional gene replacement approaches such as insertional mutagenesis and lack of endogenous regulatory control of expression. We have recently reported that zinc finger nuclease (ZFN) driven gene correction can be achieved in vivo in a neonatal mouse model of hemophilia by combining AAV-mediated delivery of both the ZFNs and a Factor IX donor template with homology to the targeted F.IX gene (Li et al., Nature, 2011). The mouse model carries a mutant human F.IX mini-gene (hF9mut) knocked into the ROSA26 locus and ZFN-mediated cleavage followed by donor-dependent repair results in restoration of functional F.IX expression. AAV-ZFN and AAV-Donor vectors were administered to neonatal mice, where the rapid proliferation of hepatocytes in the growing animal may promote genome editing through homology directed repair (HDR). Here we sought to investigate whether ZFN-mediated genome editing is feasible in adult animals with predominantly quiescent hepatocytes. Tail vein injection of the AAV-ZFN and AAV-Donor, containing a promoterless wild type factor IX insert flanked by arms of homology to the target site, into adult (8 week old) mice (n=17) resulted in stable (>10wk) circulating F.IX levels of 730–1900 ng/mL (15-38% of normal), whereas mice receiving ZFN alone (n=9) exhibited F.IX levels below detection (<15 ng/mL). Co-delivery of AAV-Mock (luciferase expressing) & AAV-Donor (n=9), yielded <65 ng/mL F.IX. Importantly, mice lacking the hF9mut gene averaged less than 100 ng/mL after receiving AAV-ZFN and AAV-Donor (n=8), suggesting that F.IX expression was derived from on-target genome editing. To eliminate the potential for hF.IX expression resulting from episomal (non-integrated) AAV genomes we performed a two-thirds partial hepatectomy two days after AAV administration. Liver regeneration following hepatectomy is known to substantially reduce expression from non-integrated AAV genomes yet no significant differences in transgene expression were observed compared to non-hepatectomized mice: circulating F.IX levels in the AAV-ZFN + AAV-Donor group (n=13) ranged between 678–1240 ng/mL, whereas mice receiving ZFN alone (n=8) or Mock + AAV-Donor (n=8) had no detectable F.IX expression, or <100 ng/mL F.IX, respectively. Taken together, these data suggest that the F.IX expression in ZFN + Donor treated mice was derived from stable correction of the genome at the intended target site.

In summary, we have shown that synchronized cell proliferation of hepatocytes, either in neonatal mice or following partial hepatectomy, is not necessary to achieve highly efficient genome editing and resultant high levels of transgene expression in vivo. These findings substantially expand the potential of ZFN-mediated genome editing as a therapeutic modality.