Adeno-associated virus (AAV) vector structure. The transgene expression cassette is flanked by AAV2 inverted terminal repeats (ITR). All plasmids used contained stuffer and Simian vacuolating virus 40 (SV40)-derived polyA insulator sequences followed by the albumin enhancer-a1-antitrypsin promoter (EalbAATp). Either human alanine-glyoxylate aminotransferase (AGXT), including its 5′ untranslated region, or enhanced green fluorescent protein-complementary DNA (GFP cDNAs) were used, followed by a woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) only in those constructs for AAV8 production, and the bovine growth hormone polyadenylation sequence. Thus, the four AAV plasmids used were ssAAV-EalbAAT-AGXT-WPRE-polyA, ssAAV-EalbAAT-AGXT-polyA (expressing the therapeutic gene), ssAAV-EalbAAT-GFP-WPRE-polyA and ssAAV-EalbAAT-GFP-polyA (expressing the reporter gene GFP).

Changes in 24-hour oxalate excretion (oxaluria) in a mouse genetic model for primary hyperoxaluria type I (Agxt1KO) after gene therapy. (a) Oxaluria in Agxt1KO mice treated with AAV8. Marked reduction in oxalate excretion is observed in adeno-associated virus 8–alanine-glyoxylate aminotransferase (AAV8-AGXT) treated mice (AGXT cluster of bars) compared with controls [green fluorescent protein (GFP) cluster], already evident at 1 week. Four weeks after injection, AAV8-AGXT treated mice had oxaluria levels not significantly different from the wild-type (Agxt^+/+) group (WT). (b) Increase in oxalate excretion after ethylene glycol (EG) gavage. AAV8-AGXT treated mice (AGXT) responded with a mild oxaluria increase (0.34±0.37 µmol/24 hour), while controls (GFP) increased oxalate excretion an average 1.24±0.57 µmol/24 hour (c) Increase in oxalate excretion with 0.5% EG in drinking water. AAV8-AGXT treated mice showed a functional reserve even better than the wild-type mice (WT), resulting in oxaluria elevations of only 0.18±0.34 µmol/24 hour, while the GFP control group increased oxalate excretion by an average 2.65±1.31 µmol/24 hour (d) Increase in oxalate excretion with 0.7% EG in drinking water. AAV8-AGXT treated mice showed an increase in urine oxalate lower than wild-type animals (WT), while the GFP control group responded with a dramatic oxaluria increase. Three animals developed renal failure and died in the GFP control group, while none of the AAV8-AGXT treated mice died. Bars: mean values; Error bars: standard deviation of the means. Nonparametric tests: Mann–Whitney for two unpaired groups, Kruskal–Wallis for three unpaired groups and Friedman test for several repeated measures.

Changes in 24-hour oxalate excretion (oxaluria) in a mouse genetic model for primary hyperoxaluria type I (Agxt1KO) after gene therapy with adeno-associated virus 5 (AAV5) vectors. (a) Oxaluria in male Agxt1KO mice treated with AAV5–alanine-glyoxylate aminotransferase (AGXT). A marked reduction in oxalate excretion is observed in mice treated with either dose of adeno-associated virus 5–alanine-glyoxylate aminotransferase (AAV5-AGXT) in males, compared with AAV5-green fluorescent protein (GFP) controls. (b) Oxaluria in femaleAgxt1KO mice treated with AAV5-AGXT. Only females injected with either 1.5×10¹³ vg/kg or 5×10¹² vg/kg but not 5×10¹¹ vg/kg showed a significant decrease in oxalate excretion, compared with AAV5-GFP controls. Bars: mean values; Error bars: standard deviation of the means. Nonparametric tests: Friedman test for repeated measures.

Alanine-glyoxylate aminotransferase (Agxt) expression (western blot) in liver of adeno-associated virus (AAV)-AGXT treated mice. Fifty µg liver protein from Agxt^−/− mice treated with either 5×10¹² vg/kg AAV8-AGXT, 5×10¹² vg/kg AAV5-AGXT or 5×10¹¹ vg/kg AAV5-AGXT were probed with affinity-purified rabbit antibody raised against recombinant mouse AGT and reprobed with rabbit anti-Gapdh serum as a loading control. High levels of AGT protein, higher than those observed in wild-type mice (wt), are seen in both male (upper panel) and female (lower panel) mice injected with 5×10¹² vg/kg AAV8-AGXT. The same dose of AAV5-AGXT virus resulted in robust Agxt expression in males, also higher than in wt mice, but significantly lower levels of expression were seen in females. Injection of 5×10¹¹ vg/kg AAV5-AGXT resulted in lower Agxt expression in males, while it was detectable only after longer exposures in females. Three representative samples from each group are presented. Five minutes exposure.

Kidney histology and liver immunohistochemical detection of alanine-glyoxylate aminotransferase (Agxt). (a) Severe nephrocalcinosis in Agxt^−/− mouse treated with control AAV8-green fluorescent protein (GFP), while AAV8-AGXT treated animal shows no calcium oxalate (CaOx) deposits. Bar = 250 µm. (b) Immunohistochemical staining for AGT. Adeno-associated virus 8 (AAV8)-AGXT treated Agxt^−/− mice shows high levels of expression while no immunostaining is seen in the liver of mice injected with AAV8-GFP. Slides were scanned together to show the intense, homogeneous expression achieved over the entire liver lobule, at the end of the study (8 weeks after injection). (c) Subcellular localization of AGT protein. Agxt^−/− mice injected with AAV8-AGXT showed abundant punctated AGT signals, which colocalize with peroxisomal marker PMP70 and not with mitochondria. Bar = 25 µm. (d) AGT immunohistochemistry on male and female livers treated with AAV8-AGXT and AAV5-AGXT. Most of the hepatocytes from male mice injected with 5×10¹² vector gemomes (vg)/kg of either AAV8 or AAV5-AGXT are positive. Lower percentages of hepatocytes were detected in female mice treated with the higher dose of AAV5-AGXT. After injection of 5×10¹¹ vg/kg AAV5-AGXT, around 15% hepatocytes showed AGXT immunostaining in males, while the percentage of transduced hepatocytes with this dose was around 6% in females. Bar = 100 µm.