Livers from animals administered with AAV were collected at P90 and levels of transgene expression were assessed by both analysis of eGFP-positive cells on cryo-sections (Fig. 1A and Fig. Afatinib clinical S1A) and anti-eGFP Western blot (Fig. 1B, Fig. S1B and Fig. S2A). We observed a higher number of eGFP-expressing cells (Fig. 1A, right panels and Fig. S1A) and a stronger (but not significantly different) intensity of eGFP bands (Fig. 1B, P90 white and grey bars, Fig. S1B and Fig. S2A) in livers from animals injected at P30 compared to those injected at P4. Concordantly, the number of AAV genome copies/molecule of diploid genome (gc/mdg) was significantly higher in livers from animals injected at P30 than in those injected at P4 (Fig. 1C, P90 white and grey bars).

Figure 1 AAV2/8 administration to newborn rats is associated with vector genome dilution. To confirm that the reduced amount of vector genomes and the lower transgene expression levels observed in livers of rats injected at P4, as opposed to those injected at P30, were due to the proliferation-related dilution of episomal AAV vector genomes, we injected rats at P4 with AAV2/8-TBG-eGFP and collected livers at P15, P30 and P90. We observed the highest number of eGFP-positive cells (Fig. 1A), highest eGFP expression levels (Fig. 1B and Fig. S2B) and vector gc/mdg (Fig. 1C) in livers collected at P15. These were reduced in animals sacrificed at later time points with the strongest reduction observed between P15 and P30.

In addition, the eGFP-positive cells in livers collected at P90 appeared in clusters when rats were injected at P4 but not at P30 (data not shown), possibly representing clones from a single cell containing integrated vector genomes, as previously described in mice injected at birth with AAV2/8 [30]. Thus, as previously reported [9], [30], [31], hepatocyte proliferation results in dilution of AAV vector genomes in transduced cells and reduced efficacy of long-term liver transduction in newborn rats. To test if AAV vector genome dilution occurs in liver of adult animals, we injected rats with AAV2/8-TBG-eGFP at P30 and collected their livers at P41 and at P90. Western blot analysis showed in some but not all the livers collected at P41 slightly higher levels of eGFP expression than at P90 (Fig. S2C). However the quantification Batimastat of the corresponding eGFP band intensity did not evidence statistically significant differences (Fig. 1B, white and dark grey bars). Thus, although we can’t exclude that in some animals vector dilution occurs in liver of rats injected at P30, overall this was not significant. Interestingly, the amount of liver vector genomes at P41 is significantly higher than at P90 (Fig. 1C, white and dark grey bars).